Skip to main content

Variables in Shell Scripting

Variables are used to store data, such as numbers, text strings, or command outputs, so they can be referenced and manipulated throughout your script.

Defining Variables​

In Bash, you define a variable without any spaces around the equals sign =.

#!/bin/bash

# Correct variable definition
NAME="John Doe"
AGE=30

# Incorrect (will cause an error)
# NAME = "John Doe"

Accessing Variables​

To access the value stored in a variable, prefix the variable name with a dollar sign $.

echo "My name is $NAME and I am $AGE years old."

For better safety, especially when concatenating strings, you can use curly braces inside code blocks:

WORD="script"
echo "I am writing multiple ${WORD}s."

Note: In the markdown text of these docs, we avoid using curly braces outside code blocks, but they are essential inside Bash scripts for variable expansion!

Types of Variables​

  1. Local Variables: Created within a script and only available within that script.
  2. Environment Variables: Available to the shell and any child processes (e.g., $PATH, $USER, $HOME).
  3. Special Variables: Built-in variables holding specific state data.
    • $0 - The name of the script.
    • $1 to $9 - The first 9 arguments to the script.
    • $# - The number of arguments passed to the script.
    • $? - The exit status of the last executed command.

Command Substitution​

You can store the output of a command in a variable using command substitution $() or backticks.

CURRENT_DATE=$(date +%Y-%m-%d)
echo "Today is $CURRENT_DATE"

Best Practices​

  • UPPERCASE vs lowercase: Conventionally, environment variables are uppercase (PATH), and script-local variables are lowercase (my_var).
  • Quoting: Always quote your variables when using them (e.g., "$MY_VAR") to prevent word splitting and globbing issues.
  • Read-only Variables: Use the readonly command for variables that shouldn't change.
readonly PI=3.14159