Introduction to Unix

../_images/ksus5.jpg ../_images/UNIX_blocks5.jpeg

4.12. Shell Variables

Several system defined variables are set for you when you log in.

Name Meaning
$HOME Absolute pathname to your home directory.
$PATH List of directories to search for commands.
$USER Your user name.
$SHELL Absolute pathname of your login shell.
$TERM The type of your terminal.

You may also define new variables within your shell or shell script program. Variables to the shell are created using a simple assignment statement with no spaces:

variableName=value

me=$(whoami)

echo "Hi, my name is $(getent passwd $me | cut -d: -f5)."

Note

  • Variables are by default strings. Holding numbers in variables and math on variables is covered later. (see Math in Shell Scripts)
  • When assigning a value to a variable, do not use the $. But use the $ when assessing the variable.

4.12.1. export

export

NAMEs are marked for automatic export to the environment of subsequently executed commands.

SYNOPSIS

export [name[=value] …]

Existing variables may be exported. Variables may also exported as their value is assigned.

me=$(whoami)

export me

export today=$(date)

See also

source

4.12.2. env

env

Display a list of the shell environment variables or run a command with modified environment variables.

SYNOPSIS

env

env [OPTION]… [-] [NAME=VALUE]… COMMAND [ARG]…

OPTIONS

-i, --ignore-environment

start with an empty environment. A mere - implies -i.

-u, --unset=NAME

remove variable from the environment before running the command.

If no COMMAND, print the current environment variables.

4.12.3. Command Line Arguments

Arguments or variables may be passed to a shell script. Simply list the arguments on the command line when running a shell script. In the shell script, $0 is the name of the command run (usually the name of the shell script file); $1 is the first argument, $2 is the second argument, $3 is the third argument, etc…

Now for something subtle…

The commands echo $* and echo $@ both print the same thing, the list of all command line arguments, but “$*” is one string, and “$@” is a list of separate strings for each parameter. Consider the following two examples to see how these two similar variables differ. In both cases, the shell script file is ran with three simple command line arguments as ./myscript a b c.

for i in "$*"
do
   echo $i
done

Results in:

a b c

Now here is:

for i in "$@"
do
   echo $i
done

which displays:

a
b
c

Note

The variable $# reports the number of command line arguments passed to the shell script program.

4.12.4. shift

shift

Cause all of the positional parameters $2 to $n to be renamed $1 to $(n-1), and $1 to be lost.

SYNOPSIS

shift

Here is an example that just displays all of the command line arguments:

while [ ! -z $1 ]
do
    echo $1
    shift
done

Note

Testing of string variables and shell script control constructs are covered later.

4.12.5. Exit Code

Linux for Programmers and Users, Section 5.21.

When a program is ran, the success or failure of the program may be determined by evaluating the variable $?. An exit value of 0 (zero) means the program was successful, and a nonzero exit value (usually 1) indicates failure.

exit

Return an exit value from a shell script.

SYNOPSIS

exit [n]

Note

When writing programs in C for Unix, it is customary for main() to return an integer (int) value. Zero (0) for success and 1 for failure.