Get started with bash script

A great feature of the Linux shell is its programming capability. This feature provides feasibility of managing complex computations. This session focuses on the basic of the bash script. You will learn how to compose a simple bash script, make the script executable and run it as a shell command.

The first script in action

Follow the steps below to write our first bash script, and put it in action.

  • Change present working directory to $HOME/tutorial/libs

    $ cd $HOME/tutorial/libs

  • Create a new text file called hello_me.sh

    $ nano hello_me.sh

  • Save the following texts into the file

     1#!/bin/bash
 2
 3# The -n option of echo command do not print the new line character at the end,
 4# making the output from the next command to show on the same line.
 5echo -n "Hello!  "
 6
 7# Just run a system command and let the output printed to the screen
 8whoami
 9
10# Here we capture the output of the command "/bin/hostname",
11# assigning it to a new variable called "server".
12server=$(/bin/hostname)
13
14# Here we compose a text message and assign it to another variable called "msg".
15msg="Welcome to $server"
16
17# Print the value of the variable "msg" to the terminal.
18echo $msg

  • Change the file permission to executable

    $ chmod a+x hello_me.sh

  • Run the script as a command-line tool

    $ ./hello_me.sh

    Note

    In addtion to just typing the script name in the terminal, we add ./ in front. This enforces the system to load the executable (i.e. the script) right from the present working directory.

Interpreter directive

Generally speaking, a shell script is essentially a text file starting with an interpreter directive. The interpreter directive specifies which interpreter program should be used to translate the file contents into instructions executed by the system.

The directive always starts with #! (a number sign and an exclamation mark) followed by the path to the executable of the interpreter. Since we are going to use the interpreter of the bash shell, the executable of it is /bin/bash.

Comments

Except for the first line that is meant for the interpreter directive, texts following a # (number sign) in the same line are treated as comments. They will be ignored by the interpreter while executing the script. In BASH, there is no special syntax for block comments.

Shell commands

Running shell commands via a script is as simple as typing the commands into the text file, just like they are in the termianl. A trivial example is show on line 8 where the command whoami is called to get the user id.

Variables

Variables are used to store data in the script. This is done by assigning value to variable. Two different ways are shown in the example script:

  1. The first way is shown on line 12 where the variable server is given a value captured from the output of the /bin/hostname command. For capturing the command output, the command is enclosed by a parenthesis () following the dollar sign $.

  2. The second way shown on line 15 is simply assigning a string to the variable msg.

Note

When assigning value to variable, there SHOULD NOT be any space characters around the equal sign =.

Tip

Environment variables are also accessible in the script. For example, one can use $HOME in the script to get the path to the personal directory on the file system.

Note

BASH variables are type-free, meaning that you can store any type of data, such as a string, a number or an array of data, to a variable without declaring the type of it in advance.

This feature results in speedy coding and enables flexibility in recycling variable names; but it can also lead to conflict/confusion at some point. Keep this feature in your mind when writing a complex code.