The shell and the terminal

A shell is a user interface for access to an operating system’s services. Most often the user interacts with the shell using a command-line interface (CLI). The terminal is a program that opens a graphical window and lets you interact with the shell.

Background

Originally, a computer terminal was an electronic or electromechanical hardware device used for entering data into, and displaying data from, a computer or a computing system. The terminal of the first working programmable, fully automatic digital Turing-complete computer, the Z3 (1941), had a keyboard and a row of lamps to show results.¹

Early computers where huge machines taking up a lot of space. Commonly a system consisted of multiple cabinets, for example one cabinet for the main processor unit, one or more cabinets for tape drives, one cabinet for each disk drive, one cabinet for a punched card reader and one cabinet for a high speed printer. In the below image, a Univac 9400 system (1967) consisting of multiple cabinets is shown.

Teletypewriter (TTY)

Early user terminals connected to computers were electromechanical teleprinters or teletypewriters (TeleTYpewriter, TTY). In the above image of the Univac 9400 system, the cabinet marked UNICAC 9400 is the main processor cabinet. The terminal is the machine looking like a huge typewriter placed on the desk to the left of the main processor cabinet. Another example of an early terminal is the Teletype Model 33 ASR (1963) shown below.

Video display terminal

As technology improved, teleprinter terminals was replaced by video display terminals. One example of such a video display terminal is the DEC VT100 (1978) shown below.

Note that the DEC VT100 terminal shown above is not a computer. The DEC VT100 terminal was only used for input and output to and from a connected computer. In the below image DEC VT52 video terminal (1974) is connected to a PDP 11/55 computer (1975).

Terminal emulator

A terminal emulator is a program that emulates a video terminal within some other display architecture.² Today, the term terminal is often used synonymously with a terminal emulator running a shell.

The department Linux system

Here you find information on how to log in to the University’s Linux system from one of the computer rooms for students on Campus Ångströmlaboratoriet.

Windows in all computer rooms

In all computer rooms on campus Ångstrumslaboratoriet you find computers with Windows.

ThinLinc

On campus Ångströmlaboratoiet there are a number of large computers (servers) running Linux somewhere in a basement. As a student, you will never physically interact with these server computers. Instead, you log in to these servers from Windows to get access to the Linux desktop environment on the same screen as Windows. The system used for this is called ThinLinc.

Log in to Windows

To log in to Windows, enter the username of your student account on the form abcd1234 and Password A.

Software Center (ZENworks)

Somewhere on the Windows desktop you will find Software center (Zenworks).

Double-click the Software center (Zenworks) icon. Now a new windows opens with available software.

The software you find here may already be installed on the computer. If a program is not already installed, it will be installed the first time you trying to use the program.

Start the ThinLinc client

From Software Center, find the ThinLinc client.

Double-click on ThinLinc Client. If ThinLinc Client is already installed , it will start quite quickly. If not installed already, ThinLInc Client will first be installed and then started.

The Windows taskbar

Once the ThinLinc client started, the ThinLinc client icon will appear in the Windows taskbar.

Log in to Linux with the ThinLinc Client

Click on the ThinLinc Client icon in the taskbar to log in to the Linux system. Enter the following login information.

• Server: thinlinc.student.it.uu.se.
• Username: Your student account username on the form abcd1234.
• Password: Your Password A.

Click Connect to login. The first time you log in, you may see the following message.

If you see the above message, click on Continue.

The Linux desktop

After a successful login, new window will open with the Linux desktop environment. You can maximize or resize this window, just like for any other window in Windows. Thus, in the Linux desktop window in Windows you interact with the Linux system that is actually running on a remote server.

The department Linux system runs Ubuntu Server 18.04 LTS using the Gnome flashback desktop environment with the Numix desktop theme.

The Gnome terminal

You can open a terminal in two ways:

• From the Applicatinos menu at the top left of the desktop: Applications → Accessories → Terminal.
• By pressing the keyboard shorcut CTRL + ALT + T .

After a few seconds a new terminal window should open.

In the upper left corner of the white area of the terminal window you see abcd1234@arrhenius:~$ . This it the shell prompt with your username on the form abcd1234 and the name of the Linux server you are connected to, in this example arrhenius. The shell prompt you see might be different.

The shell prompt

In the above example, the prompt shows the username of the logged in user abcd1234 together with the name of the physical Linux server arrhenius used. You should see your own user name. If you are logged into a different physical Linux server you will also see a different server name in the prompt.

It is also possible to tweak the prompt to show custom information such as your username, local time etc.

No shell prompt in instructions

Since the appearance of the shell prompt might vary, in all further instructions the shell prompt will be omitted and commands you enter att the shell prompt will be presented in a box like this.

ls -F  # Example shell command

If you hover over the box above, a copy button will appear in the upper right corner. Press this button to copy the text in the box. Now you can paste the copied command at the shell prompt in your terminal and press enter to execute the command.

Log out from the Linux system

There are two options for logging out from the Linux system. To log out of Linux you can:

1. Click on the icon with a person walking through a door at the top.

2. Click on the icon with a computer at the top right and then select Log Out to log out.

After you log out, the Linux window closes. If you logged out by mistake you must restart the ThinLinc Client and login again to continue working with Linux.

Remote login to the department Linux system with SSH

You can access the [department Linux System ][dep-linux] remotely using SSH to login to one of the student servers.

• You will not be able to access the graphical desktop environment.
• But, you can start graphical applications from the command line (shell) if you use X forwarding together with SSH.

In the below example, a user with user name abcd1234 uses the ssh command with the -X option to enable X forwarding to log in to the department Linux server trygger.it.uu.se.

ssh -X abcd1234@trygger.it.uu.se

One you are logged in, you can start graphical applications from the Linux shell. You can for example run Mars.

mars

To edit C source code you can for example use the VS Code source code editor remotely.

code

Known problems

A list of know problems with the department Linux system and workarounds.

Working in the terminal

On this page your find a small collection of useful shell commands when working in the terminal. You will also learn about shell variables, the command history and how to read manual pages in the terminal.

Your username (whoami)

Every user on the Linux system has a unique username. The whoami command will show your username. Type whoami at the shell prompt.

whoami

Press enter to execute the command. Now the result will be printed on the next line in the terminal and a new shell prompt will appear on the line after that.

abcd1234

In the above example the username of the logged in user abcd234 is printed as the result of the whoami command.

Print working directory (pwd)

The shell has a concept of a current working directory. The pwd (print working directory) commands prints the full path of the current working directory.

Type pwd at the shell prompt.

pwd

Press enter to execute the command.

/home/abcd1234

In the above example the current working directory /home/abcd1234 is printed as the result of the pwd command.

Home directory

On the Linux system each user has a private home directory to where she/he can save files and create sub directories.

When you first log in to the Linux system the home directory will be used as the current working directory in the shell.

For user abcd1234 the full path to the home directory is /home/abcd1234.

List files and directories (ls)

To list the files and directories in the current working directory the ls command can be used. The name ls is a short form of list (files).

Type ls at the shell prompt.

ls

Press enter to execute the ls command. You should see something similar to the below as result but you might see other files and folders listed.

foo.txt  Desktop  public_html

In the above example the only content in the current working directory is the text file foo.txt and two sub directories Desktop and public_html. You may see many more directories and files.

Distinguish between files and folders (ls -F)

To get some more information about files and folder various options can be given to the ls command. One useful option is -F that marks directories with a trailing slash /.

ls -F

You should now see something similar to this.

foo.txt  Desktop/  public_html/

Visualize a directory as a tree

The tree command displays the contents of the current directory and subdirectories as a tree structure.

tree

The output takes a graphical form which will resemble the following example:

.
├── README.md
├── one.txt
├── sub
│   └── three.txt
└── two.txt

1 directory, 4 files

In the above example, there are three files (README.md, one.txt and two.txt) and one sub directory (sub) in the current working directory. In the sub directory sub there is a single file three.txt.

You can provide tree with the path to a directory to visualize its content.

tree sub

Now, only subtree of the sub directory is shown.

sub/
└── three.txt

0 directories, 1 file

brew install tree

Change directory (cd)

The cd command navigates to a different folder. The name cd means change directory.

First print the current working directory.

pwd

This will show the path of the current working directory, for example.

/home/abcd1234

To navigate to the Desktop folder, type cd Desktop at the shell prompt and press enter.

cd Desktop

Print the current working directory to confirm.

pwd

You should now see your absolute path to your Desktop directory.

/home/abcd1234/Desktop

Note how the current working directory changed from /home/abcd1234 to /home/abcd1234/Desktop as the result of the cd Desktop command.

The directory above the current working directory can be referred to using ... To navigate to the parent directory, type cd .. and press enter.

cd ..

Now, execute the pwd command again.

pwd

And, you are back in your home directory.

/home/abcd1234

Note how the current working directory changed back from /home/abcd1234/Desktop to /home/abcd1234 as the result of the cd .. command.

Print content of file to the terminal (cat)

The cat command can be used to print the content of a file to the terminal.

Assume you have the following file named foo.txt in the current working directory.

The first line of the file.

The third line. The second line is empty.
The last line of the file.

You can now print the content of foo.txt to the terminal using the cat command.

cat foo.txt

Now, the content of the file foo.txt will be printed out in the terminal.

The first line of the file.

The third line. The second line is empty.
The last line of the file.

The name cat is a short form of concatenate which means to join together. If more than one argument is given to cat the contents of the provided files will be joined together and printed to the terminal.

In the below example cat is used to concatenate the file foo.txt with itself.

cat foo.txt foo.txt

This will output the contents of the file foo.txt twice.

The first line of the file.

The third line. The second line is empty.
The last line of the file.
The first line of the file.

The third line. The second line is empty.
The last line of the file.

One useful option to the cat command is -n:

cat -n foo.txt

, which prefixes each line with a line number.

     1	The first line of the file.
     2
     3	The third line. The second line is empty.
     4	The last line of the file.

Count words, lines and bytes (wc)

The wc command counts the number of words, lines and bytes.

wc foo.txt

In this example the file foo.txt has four lines of text with a total of 20 words and 98 bytes.

       4      20      98

In the above example we see that the file foo.txt contains for lines, 20 words and 98 bytes.

Filter (grep)

The grep command searches its input for a pattern and prints all lines in the input that contains that pattern.

To search for the the string X in the input type grep X at the shell prompt and press enter.

grep X

Note that we don’t get back the shell prompt. This is because the grep command is still running waiting for input. The grep command will now read input from the terminal and print back all lines containing the character X.

Now type Hello and press enter.

Hello

There is no X in the string Hello and therefore grep will not print back the string Hello to the terminal.

Type Hello mr X and press enter and watch what happens.

Hello mr X
Hello mr X

Once you type Hello mr X the grep command will print Hello mr X right back to the terminal since it contains a matching X.

Lets try a few more lines and observe what happens.

abc
abcXdef
abcXdef
xxx
xXx
xXx

Only lines containing a matching X will be echoed back to the terminal.

Press Ctrl D. Now grep terminates and you get back to the shell prompt.

To filter the lines in a file, the name of the file can be given together with a search pattern to grep.

Assume you have the file foo.txt in your current directory. Using cat:

cat foo.txt

, prints the contents of the file to the terminal:

The first line of the file.

The third line. The second line is empty.
The last line of the file.

In the below example only lines containing of in the file foo.txt will be printed to the terminal.

grep of foo.txt

The above command will result in the following being printed to the terminal.

The first line of the file.
The last line of the file.

Filter the output of ls using grep (ls | grep)

The usefulness of grep might not obvious at this point. To make grep useful we will combine grep with ls to filter the output of ls.

First we use ls to list all files and folders.

ls
foo.txt    Desktop    public_html  

If we are only interested in files (and folders) with names ending in .txt we can combine ls and grep to using the pipe character |.

ls | grep .txt

In this example, only the foo.txt files matches the .txt pattern.

foo.txt 

In the above example, first the ls command executes but it does not print its result back to the terminal. Instead, the result of the ls command becomes the input to the grep command. The only file or folder name containing .txt is foo.txt.

Compressed file archives (tarballs)

It is often useful to compress multiple files and folders into a single file that can later be decompressed and expanded to get back the original files and folders. There exists many file formats for compressed file archives.

• Windows users commonly use the zip file format.
• Unix users commonly use the tar file format.

Download the following gziped compressed tar archive (tarball) to your home folder:

• archive.tar.gz

Verify that you have the tarball in your current working directory

From the terminal, make sure you have the downloaded tarball in the current working directory. If you have many files in the current working directory you can use ls together with grep to search for files with names matching .tar.

ls | grep .tar

Hopefully you will see the downloaded tar ball in the result.

archive.tar.gz 

In the above example the output of ls is piped together with grep to filter the output of ls to only print any files (or folders) containing .tar. You should see archive.tar.gz among the results.

Sneak peek inside a tarball (tar tf)

To see the contents of a tarball without extracting all the files you can use tar with options t and f.

tar tf archive.tar.gz

In this example this is the content of the archive.tar.gz tar ball.

archive/
archive/large.txt
archive/small.txt
archive/sub_folder/
archive/sub_folder/info.txt

In the above example we see that the tarball archive.tar.gz contains the top level directory archive with sub folder sub_folder. In the top level directory archive there are two files (large.txt and small.txt) and in the sub folder sub_folder there is a single file (small.txt).

Unpack a tarball (tar xvfz)

To unpack and extract the contents of a gzipped tarball we need to use the xvfz options together with the tar command.

tar xvfz archive.tar.gz

Now the name of each directory/file that is extracted is printed to the terminal.

x archive/
x archive/large.txt
x archive/small.txt
x archive/sub_folder/
x archive/sub_folder/info.txt

Now the tarball have been unpacked. Use ls to see what happened to the current working directory.

ls | grep archive

Now you should have both the tar ball archive.tar.gz and the extracted archive in your working directory.

archive
archive.tar.gz

In the above example we now have a new directory named archive inside the current working directory.

Use cd to “step inside” the archive directory.

cd archive

Next, use ls -F to list the content in this directory.

ls -F

This is the content of the archive folder.

large.txt
small.txt
sub_folder/

Using the -R option ls will be run recursively stepping inside every sub-directory.

ls -R

The contents of the archive folder viewed recursively.

large.txt	small.txt	sub_folder

./sub_folder:
info.txt

In the result printed by ls -R a single period . means the current working directory.

Print text back to the terminal (echo)

To print anything to the terminal simply type echo followed by the text you want to print.

echo Hello

The text Hello now appears in the terminal.

Hello

Note that Hello is echoed back to the terminal as the result of executing the echo Hello command before the shell prints the next command prompt.

Shell variables

The shell can set and read variables. Sometimes it is useful to use the value of a built-in shell variable to make a command more generic and/or portable.

Remember that the command woami can be used to print your username.

whoami

In this example your username is abcd1234.

abcd1234

$USER

An alternative to woami is to use echo together with the shell variable USER. In order for echo to know if you want to print the string "USER" or the value of the shell variable USER shell variables must be prefixed with $ or enclosed within ${ }.

This:

echo Hello USER

, results in:

Hello USER

But this:

echo Hello $USER

, results in:

Hello abcd1234

And this:

echo Hello ${USER}

Results in:

Hello abcd1234

$HOME

Another useful shell variable is HOME with the full path to the home directory for the logged in user. You can use echo to check the value of the HOME variable.

echo $HOME

In this example the result is:

/home/abcd1234

Command history

Often you type and run a command in the terminal and later you wants to run the very same command again. To prevent you from having to type the same thing again the shell keeps a history of executed command. To navigate the history, simply press the up-arrow to move backwards in history and press the down-arrow to move forward in history.

Try the following command in the terminal:

pwd

, resulting in:

/home/abcd1234

And now this command:

whoami

, resulting in:

abcd1234

If you want to repeat the whoami command, simply press the up-arrow key once. Instead if you wish to run the pwd command again, press the up-arrow key twice.

Reading manual pages (man)

For more information about command you can always refer to the corresponding built in manual page. For example, to read the manual page for the ls command simply type man ls and press enter at the shell prompt.

man ls

This will print the manual one page at a time to the terminal. To view the next page, press the space bar. To quit, press q.

To learn more about the build in manual pages read the manual page about the man command.

man man

A summary of useful control keys when reading man pages.

Learn more

To learn more about the Ubuntu Linux shell:

• The official Ubuntu terminal tutorial
• Learning the Linux shell

To learn more about tar file archives (tarballs):

• The Ultimate Tar Command Tutorial with 10 Practical Examples

• The best tips & tricks for bash, explained

Git and GitHub

Source code for all tutorials and assignments will be made available in various repositories on GitHub. To download source code you will use Git to clone repositories to your Department Linux user account or to your private computer.

Check if you already have Git installed

Open a terminal, type git --version and press enter.

git --version

If git is installed you will something similar to this as a result in the terminal.

git version 1.9.1

If git is not installed you will something similar to this as a result in the terminal.

git: command not found

Install Git

If git is not installed on your system, install git by following these instructions.

Example GitHub repository

Let’s look at the following example GitHub repository.

The repository contains the folder sub and the files README.md, one.txt and two.txt. The sub folder contains a single file named third.txt. By clicking on a file, for example the one.txt file, you will see the content of the file.

A small text file.

Cloning a repository

Click on the green button named Code (1). Now a small pop-up will appear showing the repository URL. To copy the repository URL click on copy-to-clipboard icon (2) to the right of the URL.

From the terminal, navigate to a directory where you want the cloned directory to be created. To clone the example repository, type git clone followed by a space and paste the repository URL.

$ git clone https://github.com/os-assignments/git-example-repo.git

Press enter to execute the command. Now you should see something similar to this written to the terminal.

Cloning into 'git-example-repo'...
remote: Counting objects: 14, done.
remote: Compressing objects: 100% (8/8), done.
remote: Total 14 (delta 1), reused 14 (delta 1), pack-reused 0
Unpacking objects: 100% (14/14), done.
Checking connectivity... done.

Note the first line Cloning into 'git-example-repo'.... This tell you that the cloned repository is found in the newly created directory git-example-repo within the current working directory.

Investigate the cloned directory

To get an overview of the cloned repository, use the tree command.

tree git-example-repo

The tree command will print out a tree view showing all files and folders in the git-example-repo directory.

git-example-repo
├── README.md
├── one.txt
├── sub
│   └── three.txt
└── two.txt

1 directory, 4 files

From the above we see that in the git-example-repo we find the files README.md, one.txt and two.txt. In the same directory we also find a sub directory named sub. Within the sub directory we find a single file named three.txt.