The Linux read command is a bash builtin that is typically used to accept user input in a shell script. You can assign that input to a variable to be used for processing. There is however, a lot more to the read command. In this article we will take an in-depth look at the read command, it's options and show you some examples of it's usage.

Basic Usage of the Read Command

By default the read command will take input from stdin (standard input) and store it in a variable called $REPLY. The default delimiter for read is a newline, so it will accept input until you hit the enter key.

Animated gif showing the Linux read command used without any options on the command line.

Store Input in a Variable

Read always stored your input in a variable. As mentioned above it uses $REPLY for the default variable. If you want to store the input into a user defined variable, you can simply invoke the read command with a single argument. Instead of the default $REPLY variable, read will use the argument you supply. This comes in handy when writing a script that requires the user to input several different pieces of information.

Animated gif showing how to use the Linux read command to store input in a user defined variable.

Prompt User for Input

The read command comes with the -p (prompt) option, which displays a prompt to allow user input. When using the prompt option it requires text as an argument. This can be a brief explanation of what kind of input you are expecting.

For example, if you were requesting that a user input their first name, it would like like this:

read -p "Please enter your first name: "
Animated gif showing the Linux read command being used to prompt a user for input

As mentioned above, the $REPLY variable stores the input to the read command by default. If you are using this in a script and asking for multiple inputs, you can specify a user defined variable. For example, if you wanted to store the users first name in the FNAME variable, simply add the variable name to the end of the command.

[savona@putor ~]$ read -p "Please enter your first name: " FNAME
 Please enter your first name: Steve

[savona@putor ~]$ echo $FNAME
 Steve

Accepting Silent Input for Sensitive Data

You can use the read commands -s (silent mode) option to allow people to enter sensitive data. This DOES NOT secure your information in anyway other than hiding the input on the terminal. Always take special considerations when dealing with passwords in plain text.

With that being said, passing the -s option will stop the characters from being echoed to stdout (standard output).

You can use -s (silent) as a lone option, you can pair it with the -p (prompt) option and you can also specify your own variable to store it in.

Anmiated GIF showing the Linux read command using the silent option to hide input from the screen for sensitive data or passwords

Limiting User Input to x Number of Characters

You can limit the number of characters entered by using the -n (nchars) option. If you reach the number specified it will automatically stop reading input.

Animated GIF showing the Linux read command limiting input to x number of characters

As you can see in the example above, I attempted to input "okaydokay" but it automatically stopped when I hit 5 characters.

Additionally, you can use this in conjunction with the -p (prompt) option as well using a user specified variable.

Animated GIF showing the Linux read command limiting the number of characters on user prompt and placing it in a variable.

Input with Backslash

By default the backslash acts as an escape character. If you have a backslash in your input and want it to be part of the line, you can use the -r option.

In this example we first use the backslash ( \ ) then enter (escaping the new line) as a line continuation and read accepts the rest of the input. Second, we use the -r option which takes this functionality away and uses the backslash as a literal character, not it's special meaning. So when we enter backslash + enter, it ends the input.

Animated gif showing the Linux read command using -r for a literal backslash

Advanced Usage of the read Command

In this section we will show you some more advanced uses for the Linux read command. We will touch on many concepts that will help with development of scripts.

Setting a Delimiter for Input

A newline terminates the input by default. We can set this to whatever we like. For example, we can set it to a pound/hash sign (#).

Animated gif showing the Linux read command with a user specified delimiter

Creating an Array from User Input

An array holds several values, whereas a variable only holds a single value. The single value in a variable can have multiple words, but it is still treated as a single value. An array stores multiple values separately.

To store the input into an array, we can use the -a (aname) option followed by the name of the array, like so:

[savona@putor ~]$ read -a animals

This will take the input and split it into the array. Each word from the input is assigned as a value in the array, starting with 0. We can enter the values for the array separated by a space, hitting the enter key to denote the end of input.

Here is an example of putting names of animals into the ANIMALS array:

[savona@putor ~]$ read -a ANIMALS
 dog cat cow horse turtle

Now let's try to echo ANIMALS like we did earlier.

[savona@putor ~]$ echo $animals
 dog

It only printed "dog" which is the first value in the array. This is because unlike using a variable, the values in an array were split at the spaces. In order to print all the values in the ANIMAL, we have to let bash know we are using an array, not a variable.

[savona@putor ~]$ echo ${animals[@]}
 dog cat cow horse turtle

So dog is 0 in the animals array, cat is 1, and so on. We can print whichever value we like from the array using these numeric assignments.

[savona@putor ~]$ echo ${animals[0]}
 dog
[savona@putor ~]$ echo ${animals[2]}
 cow
[savona@putor ~]$ echo ${animals[4]}
 turtle

NOTE: The full use of arrays is outside the scope of this tutorial.

Rare Options of the read Command

There are a couple options that are rarely used, but we will touch on them for completeness.

Using Readline Libraries to Obtain Input

Using the -e option allows you to use readline editing in the input. The readline provides functionality to allow your to move the cursor, delete words, delete from cursor to end of line, and a lot more.

Often during an interactive session you type in a long line of text, only to notice that the first word on the line is misspelled. The Readline library gives you a set of commands for manipulating the text as you type it in, allowing you to just fix your typo, and not forcing you to retype the majority of the line. 

Gnu.org

Explaining readline is outside the scope of this tutorial, you can find links to more information below in the resources section.

Insert Text into Editing Buffer if using Readline

If you are using Readline ( -e ) then passing the -i option will insert text into the editing buffer. Here is an example:

Animated GIF showing the Linux read command used with some rare options

Conclusion

I think we covered everything there is to cover about the read command. We discussed storing input in variables, prompting for user input, creating arrays and a lot more. The read command is a great tool for all you DevOp folks out there.

Sound off in the comments below and let me know how you use the read command.

Resources and Links