Linux shell tutorial
- About the Shell.
- What is the Shell used for?
- Why is it called a "Shell"?
- How do I get into the Command-Line Shell?
- The Command Prompt.
- Entering commands.
- Clearing the screen.
- Changing directory.
- Moving files.
- Making directories.
- Removing files.
- Removing directories.
- Copying files.
- Renaming a file.
- Renaming a directory.
- Auto completion.
- Exiting the shell.
- Linux and Unix main page
About the command-line shell
The command-line shell is a text-based user interface for your operating system. Unlike a GUI shell, which uses a graphical representation of the system's services and resources, the command-line shell uses only text. It is the fundamental interface between you and the operating system, and in many cases offers you more direct control over the system processes.
What is the Shell used for?
When it comes to communicating with your operating system, anything you can do with a GUI can be done at the command line. For example, let's take something you need to do all the time: move files around in your file system. Using a GUI, you could select the icons representing your files, drag them to another folder, and drop them there to move them.
However, using the command line, this operation can be performed using a single command:
The command line, once you get used to it, allows you to control your system with greater efficiency and precision than with a GUI.
Why is it called a "Shell"?
The user interface is referred to as a "shell" because it is the outer layer separating you from the innermost parts of the operating system, called the kernel. Technically, your operating system's GUI is also a "shell," but the command line is frequently referred to casually as "the shell."
How do I get into the Command-Line Shell?
The most common way to access the command-line shell is using a terminal emulator.
If you are using a Linux distribution, it may boot into X Window System (or "X"), which allows you to run programs in their own windows and interact with them using the mouse. In X, there are different ways to launch a terminal emulator, but here are the most common:
- Many Linux distributions use a desktop environment similar to Microsoft Windows, which has a "Start Menu." From this menu, your terminal application is usually located in under "Accessories." For example, Lubuntu comes with the LXTerminal terminal emulator already installed, and the shortcut is located in the Accessories menu.
Selecting the terminal menu item launches the terminal emulator and places you at a command line.
- Different versions of Linux use different kinds of window managers, each with a slightly different interface. With many window managers, right-clicking the desktop shows a menu that contains a terminal option. For instance, the right-click menu in Crunchbang Linux has a "Terminal" option. Below is what that menu looks like, after right-clicking anywhere on the desktop.
Here, the Accessories menu would also have an option for running the terminal.
- On many versions of Linux, pressing the Super Key and "T" at the same time is the default hotkey for launching a terminal. (The Super Key is usually the first key to the left of the spacebar, also known as the Windows key).
If you cannot figure out how to launch your terminal emulator, see the documentation for your particular operating system, and search for "terminal".
The Command Prompt
When you launch the terminal emulator, you are presented with a command prompt.
Let's look at this command prompt closely. The prompt can be configured to be anything you like, but by default, it looks like what we see here, providing us with useful information before we type in a command.
Specifically, it tells us who we are (username) and where we are (what system we are using and our working directory). Here are what the individual parts of the prompt mean:
|@||An at symbol separates the username and hostname.|
|yourhost||The system hostname.|
|:||The colon is the character that separates the hostname and the working directory.|
|~||Our current ("working") directory. The tilde ("~") is a special character which represents our home directory. When you start a shell session, this is the directory where you typically begin working by default.|
|$||A special character which indicates that we do not have superuser privileges. If logged in as root (the superuser), this character changes to a #.|
Let's type in our first command. We should start by listing the files are in our home directory. The most basic way to see all of these files is by using the ls command; type ls at the command prompt and press enter.
As shown above, the ls command lists the contents of the directory and places us at another command prompt.
For a more detailed listing of the directory contents, we can specify command-line options, also known as switches. These are additional arguments to the command that modify the way it operates. Here, we will use the -l (use long list format) option to get more information from our listing.
As shown above, the ls -l command provides us with a lot more information about the contents of our home directory. At the moment, the most important information we should notice is that the first letter of the listings is a d, meaning that each of those directory contents is itself a directory.
Clearing the screen
Before we do anything else, let's clear the screen, to remove the clutter produced by the output from our previous commands. Use the clear command by typing clear at a new command prompt and press Enter.
After pressing Enter, the screen clears and places you at a new command prompt at the top of your terminal screen.
Let's move into our documents folder using the cd ("change directory") command. To change the current directory, type "cd" and then a space, then the name of the directory you would like to change to and then press enter.
The cd command assumes that the name of any directory you specify is relative to your current directory. Therefore, to change into the documents directory, which exists in our current directory, type "cd documents".
The command prompt now reflects the new working directory: ~/documents, which is equivalent to the full pathname: /home/yourname/documents. No matter what user you are logged in as, "~" always represent your home directory. In the example, the new working directory is also seen in the title bar of our terminal window.
Note: Unlike MS-DOS and Microsoft Windows, hierarchical directory names in Linux are separated by a forward slash ("/") rather than a backslash ("\"). This change of symbols is a fundamental difference between these various operating systems.
At any time, you can verify your current directory by using the pwd("print working directory") command.
The command prints the name of the current directory to the terminal, in this case "/home/yourname/documents". Again, with "~/documents" the "~" represents our home directory, which is /home/yourname. So if we are in our home directory, typing "cd documents," "cd /home/yourname/documents," or "cd ~/documents" would all place us in the same directory.
Now let's move up one directory, back into our home directory. The special directory name ".."(the double-dot) means "one directory up".
We are now back in our home directory.
Note: If you are more familiar with MS-DOS, keep in mind that there must be a space between cd and the two periods.
Now let's get familiar with how to move files around.
Let's say we just downloaded two images with a web browser, and they saved into our downloads folder. Let's go in there and take a look. First, we change directory into the downloads directory.
Then list the files in the directory with the ls command.
The mv command takes the following general form.
mv [OPTIONS...] SOURCE DESTINATION
OPTIONS is enclosed in brackets because you are not required to specify any options to complete the command. The SOURCE and DESTINATION arguments are both required.
SOURCE is the name of the file or files that you want to move. DESTINATION is where you want to move them. If DESTINATION is a directory name, the file or files are moved into that directory.
We are currently in the directory /home/yourname/downloads and we want to move the files into /home/yourname/images. We could specify our DESTINATION using the full pathname ("/home/yourname/images"), or as a pathname relative to our current directory ("../images"), or using a tilde as an abbreviation for our home directory ("~/images"). All of these are equivalent and give the same result. Let's use the relative pathname for our DESTINATION: ../images.
So, how will we specify our SOURCE?
We want to move both of the files, and they are the only files in the directory. So, the simplest way to specify them would be to tell mv "move all of the files in this directory into ../images." We can do this using a wildcard.
A wildcard is a special character which can represent more than one file name (or more than one part of the file name). Here, we are going to use the asterisk wildcard ("*"). If we use * as our entire file name, the shell applies our command to every file in our working directory.
So, our command will read as follows: "mv * ../images". Let's enter it now.
It looks like nothing has happened, but that is the shell's way of telling us no errors occurred. If there was a problem, it would display an error message or warning of some kind.
Let's verify that the files have moved. We can list the contents of another directory by running ls and specifying a pathname. In this case, our pathname is "../images."
The images are both there, along with another directory that already existed ("wallpapers"). We should make a new directory of our own, to better organize our images and keep the main images directory from getting cluttered. Let's do that now.
Let's change into the images directory where our two penguin images now reside.
Now let's make a new directory for our penguin images. Creating new directories is accomplished with the mkdir command.
The simplest form of the mkdir command is to give it one argument: the name of the new directory. We'd like to make a new directory within our current directory, so we do not have to give mkdir a full pathname, just the name of the directory itself. Let's call our new directory penguin-pix.
So, the command we are going to enter is "mkdir penguin-pix" to make a new directory.
Now let's verify it is there, by listing the directory contents.
Note: You can see that when we list the directory contents, file names and directory names are different colors to help identify the different files and directories. Although color highlighting is common with most shells, not all shells do this by default.
If we could not tell by the color of the names, we could always check which of these files are directories by invoking ls with the -l (long list) option.
The additional information includes the following.
- Permissions - who has access to the file and what kind of access.
- Ownership - who is the file's owner.
- Authorship - who created the file.
- File size - size of the file in bytes.
- mtime - date and time the file was last modified.
The first letter on the lines for penguin-pix and wallpapers is a "d," indicating that those are directories. For more information about file listings, see the documentation of the chmod command.
We can now move our images into the new directory. To tell mv what files to move, we can use a wildcard again, but this time let's make sure it knows to move only the image files, not the directories. Since both image files end with the JPEG extension .jpg, we can specify the file name "*.jpg": this tells the shell "the files I want to move are any files in the current directory that end with .jpg". So our command will look like this: "mv *.jpg penguin-pix."
No error message was given, meaning the command completed successfully. Let's verify the files were moved by listing the contents of penguin-pix.
We have two penguin pictures, but perhaps we decided that we only need real-penguin.jpg, and we do not need tux-the-penguin.jpg anymore. We can delete this file by using the rm ("remove") command.
We'll need to provide rm with the file name to delete. We are currently in the images directory, but our image is now in the penguin-pix subdirectory. We could change our working directory to penguin-pix and then run the command, but that is only extra work, we can skip that step by using the directory as part of the path.
Our command will look like this: "rm penguin-pix/tux-the-penguin.jpg."
Once again, the shell executes our command without giving any feedback. The command executed successfully, so the shell returns us to the command prompt, letting us know that it is ready for another command.
Warning: Unlike dragging a file to the trash or the Microsoft Windows recycle bin, there is no way to undo a file deletion with rm. Remove your files with care! Once gone, it is (virtually) impossible to get them back.
Let's just verify that the file is really gone.
That is great for files, but what about directories? By default, rm only removes files and not directories.
The rmdir ("remove directory") command removes a directory, assuming that it is empty. Let's test it by creating a new, empty directory, using the "mkdir test-directory" command.
Let's verify it is there with "ls -l" command.
We can now remove our brand new directory using the command "rmdir test-directory," and check that it is gone by running the "ls -l" command again.
Yup, it is gone.
However, what would happen if we tried to "rmdir penguin-pix"?
We cannot rmdir that directory because it is not empty.
If you are sure you want to remove penguin-pix and everything inside it (including all files and any directories it might contain), you can use the rm command with the -r option.
Normally, rm will not operate on a directory. However, if invoked with the -r option, rm removes a directory recursively, which means it travels into the directory and any subdirectory, deleting everything including the directory. So, if we run the command "rm -r penguin-pix," it removes the directory and the file that was inside. If there were any directories inside, they would also have been deleted.
Warning! Once again, there is no undo for this command. When you rm -r a directory, that directory and everything inside it is removed forever! Like most Linux commands, rm is very powerful and should be used with care.
cp SOURCE DESTINATION
SOURCE is the name of the file you want to copy, and DESTINATION is the pathname specifying the destination of the new file. The destination can be a new file name or the name of the directory where you would like a copy to be located.
Let's say we have a text file in our documents folder called orig.txt, and we want to make a copy of it called copy.txt. The command we should use is "cp orig.txt copy.txt."
If, on the other hand, we specify a directory as the destination of the cp command, it makes a copy with the same name as the original in the destination directory. In the next example, we do six things.
- List the contents of the working directory.
- Make a new directory, called copies.
- List the contents of the working directory, which now includes the copies directory.
- List the contents of the copies directory (there are none. It is empty).
- Copy copy.txt into the copies directory.
- List the contents of the copies directory, which now contains a copy of copy.txt.
Renaming a file
The mv command can also be used to change the name of a file. For example, the command "mv orig.txt original.txt" will rename the file orig.txt to original.txt.
Renaming a directory
Renaming directories works exactly like renaming files, with the mv command. Here, we create a directory named "copy," but rename it to "copies" using the command "mv copy copies."
If the copy directory contained any files or directories, they would be unchanged. The only thing that would change is the name of the directory that contained them.
Most Linux command-line shells automatically complete the word you have started typing if you press the Tab key. Auto-completion can help save you time, especially with long file names. For example, let's say you have three files in your documents folder: penguin-diving.jpg, penguin.jpg, and penguins.jpg.
If you type a command and then type "p" and press the Tab key the shell completes the word "penguin" for you.
It stops there because all the files start with the word penguin, but after that it is not sure which file you would like to specify.
If you are in a situation where there is more than one autocomplete match for what you have already typed, pressing Tab twice prompts the shell to list all the possible matches. So in the situation above where you typed p and Tab and the shell autocompletes penguin for you. Press Tab two more times would give you all possible options, and place you back at the prompt, with your command filled in where you left off.
Autocompletion also works for completing path names and command names.
Exiting the shell
Typing the exit command exits the command shell, and either close the terminal window, or prepare the terminal window to be closed manually.