TP n°1 Unix

Introduction

To practice on the Unix shell

You should have access to the Ubuntu session on the computer in the room, with your student number and student password. If you have a Unix-like OS (e.g., MacOS, GNU/Linux) on your laptop, you can also use it if you prefer. If you have Windows installed on your machine, you still have multiple options if you want to work on your machine. I listed some of these options on https://s.ortion.fr/w/course/mssb-computer-refresher/introduction.html

To take some notes

You should keep track of the commands you used for this practical exercise. You should avoid using a text processing application, such as LibreOffice or Word, because it might introduce unwanted character in the commands. Prefer using a plain text file, or, even better: use a standard plain text format for your notes such as a Markdown file or org-mode for instance.

Exercise 1. A first step in the Unix world

Check your unix username in two different ways
1. Use the command whoami
2. Check the content of the environment variable $USER
Create a folder with mkdir in your home directory for this practical exercise.
Move to this newly created directory with the cd command
Check your current working directory with pwd
Your Unix machine probably have the curl or wget command line included. This subject document should be available online at https://s.ortion.fr/w/course/mssb-computer-refresher/unix/tp/tp1/tp1.org and https://s.ortion.fr/w/course/mssb-computer-refresher/unix/tp/tp1/tp1.pdf if you prefer to get a PDF file. Using wget or curl, download this document in your local folder.
Using ls, check that this is correctly downloaded and present in your current directory.
Open the file you downloaded from the terminal. If you chose to download the PDF rather than the Org-mode plain text source, you will need to use a PDF viewer.
Using grep and a regular expression count how many questions there are in the tp1.org file. Each question has a number. The line of a question hence starts with the question number followed by a dot.

Exercise 2. A grep on grape

Not a long time ago took place the grape harvest. France is one of the main grape and wine producers worldwide.

The FranceAgriMer institution is a national organisation responsible of data analysis on French agricultural activities, and to apply the European common agricultural policy in France. We will use a reference document listing the grape varieties that we can find in the market. Unfortunately the reference table is stored in a inconvenient PDF format. To make your life easier, I extracted the first 100 lines of this table into a plain text file in the comma-separated values (CSV) format and translated the comment and table header: ./data/agrimer_grape.csv

source

Remember that with grep you can search for a pattern in a text file with grep PATTERN FILE. For instance, I could write the following command to find all occurrences of the term Chasselas (a well-known grape variety) in the file:
```
grep Chasselas ./data/agrimer_grape.csv
```
I can also count how many lines have this pattern term with the grep option -c:
```
grep -c Chasselas ./data/agrimer_grape.csv
```
```
2
```
I kept the PDF file header in the csv in a “comment” section, starting with the character ’#’. With grep you can also show all lines that don’t match the pattern (we sort of invert the grep query, with -v).

How would you filter out the comment lines from the csv file with a grep command ?
If you used only “#” as a grep pattern, think of what would occur if a record line contained a hashtag.

With the grep option -E you can specify a more advanced regular expression pattern (-E means extended regular expression support). What regular expression would match only the lines that starts with the hashtag ’#’ character ?

Use this to fix your previous command.
Count how many grape varieties are maintained by an institution in France (FR).
Check whether there is a grape variety that is not maintained in France.
In fact, while manually editing the file, I messed up with some lines… Those lines does not have the expected number of columns (4). Can you find these lines?

Exercise 3. An executable bash script

Open your favorite text editor (not Word, but a plain text editor), and create a file with the extension .sh.
Add a shebang in the header of this script to use the bash shell interpreter.
The file /proc/cpuinfo contains information on the CPU of your machine. Add a command to extract the model name of your CPU from this file.
Add a command to count how many available CPU of this kind there are on your system.
Make your script executable for your user and group.
Execute your script
Add the folder that contains your script to the PATH variable.
Check that you can execute your script as a command, from any folder.

Exercise 4. Getting help

A lot of information is available on the web on Linux and other Unix-like systems. Most of command line tools comes with a --help option to show a description of the options available, e.g.

grep --help

Most of standard Unix commands also comes with an entry in the Manual. To access the manual from the terminal, is often as easy as typing man <the name of the command>, e.g.,

man grep

Check the manual entry of cp. What cp option can we use to recursively copy the content of a directory? To quit the manual, type q.

Exercise 5. Internet and SSH

Can you find the ip address of your current machine from the terminal?
Knowing your Unix username and the address of your machine, and assuming that a SSH server is running on this machine, what would be the SSH command you would use to open a shell from a remote client to this machine?