adventures with microcomputers prototyping with raspberry pi

2. The Command Line

Navigate to the Command Line on Mac OSX (Applications > Utilities > Terminal).

You should see something that looks like the window with a white background at the top of this page. That is the Terminal, a name for a command line interface that is already installed on your computer. Though other aspects of computers have changed drastically over the last fifty years, this has not. In other words, you can see back in time, in a sense, at this level of your computer. Read more about the history of the command line here.

In the above photo you will notice a semi-transparent black window that looks fairly similar to the Terminal. This is iTerm2 and I prefer it to the Mac OSX Terminal. I do not really care which you use (it's a matter of personal preference), and not particularly interesting to talk about right now, just be aware that the same commands work across all CLIs.

  1. Open a new terminal window with COMMAND-N. You should see something like [HOSTNAME]:~ [USERNAME]$ followed by a rectangle. The rectangle is your cursor and demarcates where you can enter text commands for the Terminal to execute.

  2. Unix commands, and really anything used in the command line, should be shipped with some sort of manual. Typically this is accessed by typing man [NAME OF COMMAND] in the Terminal and hitting enter. Let's start by checking the manual for the next command we will use, ls. Type man ls and hit enter. You should see something that looks like this:

    This is the manual for ls, which opens automatically in a text editor called Vi (also known as Vim, short for Vi Improved). Vim has a radically different approach to how a text editor works than Microsoft Word or Apple Pages, which you can read about in detail here.

    There are two modes in Vim: Normal and Insert, and toggling between the two modes exposes different features of the application. Normal is typically reserved for navigating through a file or reading. To navigate the file you can use the j (down) and k (up) keys. In order to quit a file you simply type q.

    We will discuss how to use Insert mode, and other commands, in the future. Right now its more important to take some time to read through the manual for ls. It is considered standard practice for developers to include a man file for virtually everything that you use in the terminal, making man a very important command to commit to memory.

    Note: google vim vs emacs to see an example of how passionate people get about their tools (these arguments are frequently referred to as "religious").

  3. Type ls and then hit ENTER. ls lists the contents of the directory you are currently in. When you open a new Terminal window you will default to your user's home directory. We can confirm this by running the command pwd (print working directory), which will return /Users/[USERNAME].

  4. To move to a different directory, type cd at the command line and then the name of the directory. Terminal has autocomplete, so you could type the first two characters of the directory, for example, and then hit the TAB key, which is less work (yay!). On my machine I can move to a directory named git by running the following command: cd git/ where cd means change directory to the directory named git.

  5. You can move around on the command line with arrow keys, but there are also shortcuts like CTL-A to go to the beginning of the line and CTL-E to go to the end of the line. If you make a bunch of mistakes and just want to start over you can issue a KeyboardInterrupt by hitting CTL-C, which will result in a new line.

  6. Check to make sure you are in your home directory (pwd) and then make a new directory by typing mkdir NEWDIRECTORY, where NEWDIRECTORY is the name of your folder (name it whatever you want).

  7. In the next sequence, we will move into our NEWDIRECTORY and create a text file named test.txt:

    2. pwd
    3. touch test.txt
    4. ls
  8. Next lets write something in our text file using a text editor called Nano (which is already installed on your machine, just like Vim):

    1. nano test.txt
    2. this is my text

    Save your changes with Ctl-O, hit Enter to confirm the file name, then Ctl-X to exit.

  9. In the following sequence we will use the Unix command grep, which searches plain-text data sets for lines matching a regular expression (definition sourced from here):

    1. Click here to go to the .txt version of The Project Gutenberg EBook of Alice in Wonderland, by Lewis Carroll.
    2. Click anywhere on that page, Command-A to select all, Command-C to copy to clipboard.
    3. Back in your terminal, make a new text file called alice.txt: touch alice.txt
    4. Open the file with Nano: sudo nano alice.txt (don't forget about autocomplete with the TAB key!), you should enter your password and then hit Enter
    5. Click anywhere in the window, Command-V to paste the contents of the .txt version of Alice in Wonderland into the file, Ctl-O to save, Enter to confirm file name, Ctl-X to Exit.
    6. Now you can search all of Alice in Wonderland for every line that contains some regular expression (which you specify between quotes), which could be a word, or a string of words, but is by default case sensitive.

      Example 1: grep 'home' alice.txt results in:

      "Why, Mary Ann, what are you doing out here? Run home this moment and this creature, when I get it home?" when it grunted again so violently

      Example 2: the words 'hello', 'sorry', and 'god' do not occur in the entire text of Alice in Wonderland. In other words, grep 'god' alice.txt returns nothing.

      Note: in the future our goal would be to eliminate the steps that required the use of mouse and a web browser.

  10. Let's go ahead and delete the example files we made. rm removes files. When you need to delete a file, run rm NAMEOFFILE. If you try to use rm on a directory, like the one you made for the test.txt and alice.txt files, the Terminal will stop you (as a safety measure) because Unix thinks that directories should only be removed when they are empty. So, there are two approaches to getting around this kind of an issue:

    1. manually rm every file in the directory (this sucks because its repetitive, error-prone, and time-consuming) and then use rmdir NAMEOFDIRECTORY to remove the empty directory.
    2. ALTERNATELY, the cooler (lazier) way to do this is to use rm -r NAMEOFDIRECTORY to recursively remove all contents of the directory.

    Note: Sometimes, in order to do certain things in the terminal, you may need to prepend a command with sudo. "Sandwich," from the web comic series XKCD, sums sudo up nicely:

    sudo (short for su do) means super user do and forces the computer to let you do something. The standard practice is to avoid using sudo unless you know you have to, which often takes some guessing the first time.

    For example: I normally try a command without sudo first and, if I get a permissions error as a response, then run the following on a new line: sudo !!.

    Note: The command !!, which is pronounced bang bang, is a shortcut to repeat the previous command entered into the terminal. In other words, this command prepends sudo to whatever you ran in the terminal previously.

ls -l

The Unix command ls can do a lot more than just list the contents of a directory. If you run ls -l you can see a full list of the following information (sourced from here):

  • total files in the directory and subdirectories
  • the names of the files in the current directory
  • their permissions
  • the number of subdirectories in directories listed
  • the size of the file
  • and the date of last modification

Go ahead and try running this command in a few different directories. Try home, Applications, and a new directory, made by you (mkdir), on the Desktop expressly for this purpose.

Renaming = Moving

If you need to rename a file with Unix, use the mv command. mv renames/moves the SOURCE to the DESTINATION. For example, in the next sequence we will make a new file on the Desktop, rename it, and then move it to our home directory. Follow along in the Terminal:

  1. cd Desktop/
  2. touch test.txt
  3. mv test.txt mv_demo.txt
  4. mv mv_demo.txt ../mv_is_fun.txt Note: .. is short for up one directory level
  5. cd ~ Note: ~ is a shortcut for your home directory (so many shortcuts!)
  6. ls Note: confirms that mv_is_fun.txt has been moved/renamed

Copying Files and Directories

The cp command allows one to copy files/directories from a SOURCE to a DESTINATION. In other words, standard usage tends to look like this: cp ORIGINALFILE NEWFILE. If you need to copy a directory, though, and want to copy the entire directory structure inside of it, you can tell cp to recursively copy the directory and all of its contentscp -R OLDDIRECTORY NEWDIRECTORY.

Note: in either of the previous examples, one could also change the name of NEWFILE or NEWDIRECTORY by simply using a different name at the end of the NEWFILE/NEWDIRECTORY path.

When you need to copy files over a network you can use scp, which is virtually the same as cp but copies files or folders over a secure, encrypted network (preventing the possible interception of sensitive data via what is referred to as a Man-in-the-middle attack). You will use this command extensively when working with the Raspberry Pi 3.

Note: Though you will begin to memorize some of these commands through frequent use, it is normal to bookmark an online cheat sheet for reference.