Notes on Linux

For more background information on Linux, see the Wikipedia page at https://en.wikipedia.org/wiki/Linux. The short version: Linux is the operating system that powers my cellphone (and likely yours), many of the computers on the Web, my laptop and maybe yours too, any supercomputer anyone cares about, the LHC, and even some video game consoles.

1. Files and directories

The basic unit in Linux is the file, just like on Windows. These files reside in a hierarchical system of directories that work just like folders in other operating systems. However, instead of clicking on them, you refer to files by name, with directories separated by slashes. For instance, the file called zombie-army-roster.txt in the directory world-domination itself in the directory mad-science, which resides in my home directory /home/wfreeman would be referred to as /home/wfreeman/mad-science/world-domination/zombie-army-roster.txt. Your home directory can be abbreviated with a tilde, and all paths can be given relative to the current directory.

So, if I was currently in the directory /home/wfreeman/mad-science/, then /home/wfreeman/mad-science/world-domination/zombie-army-roster.txt is equivalent to world-domination/zombie-army-roster.txt. If you are referring to files in the current directory, you need only give their names.

Important exception: The exception is when you want to run a program: to run the program reanimate in the current directory, you must write ./reanimate. (This is for security reasons.)

File names in Linux, like everything else, are case sensitive. It’s generally a good idea to use lowercase for everything unless you have a specific reason not to. They can contain spaces, but this is a giant pain in the butt and you shouldn’t do it; use dashes or underscores instead.

2. Some basic commands

To ask Linux to do something, you use the same syntax as you would use in English for imperative sentences (commands). For instance, if I wanted you to dance, I would say “Dance!”. If I wanted to be more specific about what I wanted you to dance, I’d say “Dance the funky chicken!” I could give you details about how I wanted you to do this: “Dance the funky chicken quickly!” This can be generalized as the following:

[verb] [object] [modifiers]

Linux works the same general way: a Linux command tends to have the form

[command] [target] [modifiers]

In short, you simply type the name of the program you want to run, followed by what you want to run it on. For instance, to run Firefox, just type firefox. (Try it!) If you want a program to run with a particular target, simply say so. For instance, if I want Firefox to tell me what’s going on in the world, I could type

firefox news.bbc.co.uk.

Try this! Note that it’s running a bit slowly: this is because the Firefox window, and everything in it, is actually being run elsewhere on campus, and the graphics are being streamed to you. Years before Twitch was invented, Linux computers were happily streaming graphics to each other.

If I want to run the text editor nano to edit the file helloworld.c, I write nano helloworld.c. (You will do this later today!)

2.1 cat and less: display a file

This simply dumps the contents of a file to the screen. For instance, cat homework1.c will display the file homework1.c. Note that displaying non-text files will give you gibberish. If the file is too big for the screen, use less instead; this is just like cat, but pauses after each screenful of output.

2.2 ls: list files

ls shows a list of files in the current directory. ls -lh shows more information about them. ls -lt sorts files by how new they are. (You can combine these options; try ls -lth.)

2.3 cd and pwd: change directory, and print working directory

This changes your current working directory. Type cd .. to go up a level, or cd (with no argument) to go back to your home directory. If you ever want to know what directory you’re in, type pwd.

2.4 mkdir: create a directory

Does what it says on the tin.

2.5 rm: remove files

Removes a list of files. Warning: there is no “trash bin” on Unix like there is on other operating systems.

2.6 cp: copy files

Do cp file-one file-two to make a copy of file-one called file-two. If file-two is a directory, a copy with the same name will be created in the directory.

2.7 mv: move files

Works just like cp, but moves the file instead.

2.8 Text editors

xemacs: Graphical text editor with lots of features (somewhat beginner-friendly, powerful).

vim and its graphical cousin gvim: Another option if you dont like emacs. It is less beginner-friendly but quite powerful. (I am typing this document in vim right now.) This is what I prefer, and you will see me use it.

kate and gedit: These are more “modern” graphical editors which may run slower over a remote graphics connection from home, but provide an attractive environment for you to code in similar to the things you’re used to in Windows.

As with all programs, you can open a specific file by doing e.g. gedit helloworld.c.

2.9 g++, the C compiler

You need to use gcc, yet another Unix program, to turn the source code you write into machine code that the computer can run. To use it, do

gcc program.c -lm -o program

where program.c is the name of the file with your source code, and you want the output program (the one that you run) to be called program.

Do not make the mistake of doing -o program.c this will overwrite your source code with the output program!

Adding -lm tells the compiler to link your program with the math library, which contains useful functions that we will use frequently.

You might mix in some C++ ideas with your C code; then you will use g++ rather than gcc.

2.10 python, the Python interpreter

If you want to run Python code, there is no compiler. Instead just put your code in a file ending in .py, and say for instance python program.py. If you want to test out Python commands, you can invoke the REPL by just saying python.

3. Input and output redirection

This is the key to the versatility of the Linux command-line interface. Normally, when you run a Linux program, it takes input from the keyboard and prints output to the screen. But you can change this behavior with the <, >, and | operators.

The input redirection operator < tells a program to accept input from a file rather than from the keyboard. For instance, suppose I have a program called wharrgarbl that reads in input from the keyboard and does a calculation. If I’m getting tired of typing in all the input every time, I can put it in a text file called input, and do wharrgarbl < input.

The output redirection operator > does the same thing, but redirects the output of a program. This can be combined with input redirection. Suppose I have a program called translate that accepts typed text in one language and prints it in another. I can then use this program to translate a file by doing, for instance,

translate < sindarin-text > westron-text.

The pipe character | takes the output from one program and uses it as the input from another. Suppose I want to read in some text, translate it, and then display the output on the screen one page at a time using less. I can do this by doing translate < sindarin-text | less. The output of translate will then be sent to the standard Unix program less, which will display it one screen at a time.

4. Graphing using plot

There is a simple program called plot installed on the course computer that will make graphs for you. To use it, create a data file consisting of coordinate pairs. (Remember, you can do this using printf along with output redirection.) Then just type graph [filename]. If you have multiple files you’d like to graph, just list multiple filenames.

By default, the generated graph goes to your screen. If you’d like to save it to a PDF file instead, do

plot [filename1] [filename2] ... -o mygraph.pdf.

You may also pipe text to plot by doing something like

./myprogram | plot

or

python myprogram.py | plot

See the notes on for more information.

5. More Unix controls

5.1 Tab completion (this will save you time!)

Even if you use long filenames, you dont have to type them out. When typing out a file or directory name at the command prompt, you can type the first few letters (enough to unambiguously identify the name) and hit Tab to have the system fill in the rest. If there are several files that “match”, hitting Tab will fill in all the characters that the names share. Try it!

5.2 Task control

Sometimes you want to run one program and leave it running in the background while you do other stuff at the command prompt. This is especially useful for graphical programs (like kate), since you can edit your program in the graphical window while testing it in the shell. To do this, add an ampersand after the command:

gedit program.c &

If you forget to do this, you can suspend a running program with Ctrl-Z, then put it in the background by typing bg. This is equivalent to running it in the background in the first place. If you want to stop a program that’s running (perhaps you’ve written a program that’s stuck), use Control-C.