In-Class Review I

In-Class Review I#

This is a review of some of the concepts we covered up to this point in the semester.

Important

This set of examples is not exhaustive, so you should go back and read through the notes and run the example codes.

Bash shell#

You log into a computer–how do you find out what directory you are currently in?
solution
pwd
The pwd command stands for print working directory.
How can you return to your home directory from anywhere on the filesystem?
solution
You can simply do:
cd
cd with out any arguments always brings you home.

You could also do:
cd ~
or give cd the full / absolute path to your home directory.
What is the full command that will transfer a file from your home directory on portal to your current directory on the desktop computer you are sitting at in the MathLab?
solution
We use scp (secure copy) for this:
scp username@portal.mathlab.stonybrook.edu:~/file .
That will transfer a file called file in your home directory (~/) on portal to the current directory (.) on your local machine. You would substitute your NetID for username.
Here’s an example filesystem on a computer you are using:
```
/
├── etc
├── home
│   ├── common
│   ├── system-admin
│   ├── teacher
│   └── user
│       └── files
│           ├── report1.txt
│           ├── report2.txt
│           └── report3.txt
└── opt
```
Your home directory is user
1. What is the full (absolute) path to the file report1.txt?
  solution
  /home/user/files/report1.txt
  Remember: absolute paths begin at the root of the filesystem (/).
2. If you are in the files/ directory, how do you copy all of the files there to the common/ directory? Give a single command for this.
  solution
  We would use the copy command, cp. There are several variations of how we could specify the source files and the destination. You could do:
  cp report1.txt report2.txt report3.txt /home/common
  With cp, the last argument on the command line is always the destination. Here we explicitly write out the 3 files we wish to copy.
  
  We could also use wildcards, like:
  cp *.txt /home/common
  In this case, *.txt will match any file in our directory that ends in .txt.
3. Using wildcards, what is a single expression that matches all of the files under files/?
  solution
  In addition to *.txt that we used above, we could do any of these:
  - * : this matches every file in the directory. In this case, this is okay, since we want all the files. But usually that is not the case.
  - report*.txt : this will match any file that starts with report and has the .txt extension.
  - report?.txt : this will match any file that starts with report, then has a single character (can be a letter or a number) and then ends in .txt.
4. You want to delete the directory files/ and everything in it. How do you do this?
  solution
  Starting in your home directory, you can do:
  rm -f files
  This will recursively delete the directory, by first deleting all the files in it, and then removing the directory. This can be dangerous if you did not setup the alias for rm to always be rm -i (prompt you before deleting).
  
  Another way would be:
  cd files rm *.txt cd .. rmdir files
  This first changes directory into files, then (using a wildcard) removes all the files, then changes directory back up a level (..) and then uses rmdir, which only works on an empty directory.
You have a file on a group server (directory.txt)–how do you make sure that anyone on the machine can read it, but only you can write to it?
solution
You can do this in a few ways. One way is:
chmod a+r directory.txt chmod a-w directory.txt chmod u+w directory.txt
The first two lines set the permissions for everyone (user, group, and other), while the last line restores write permissions for the user.
What do you expect the output of the following command sequence to mean?
```
grep -i yellow database.txt | wc -l
```
solution

This will return the number of lines in database.txt that contain the word yellow, without worrying about case.
Suppose you wanted to see the last 10 commands you entered at the Bash prompt–what command sequence could do this?
solution
You could do:
history 10
or if you didn’t realize history could take a number (like me!), you could do:
history | tail -10
You have an executable in your current directory called hello. If you just do:
```
hello
```
you get an error (“command not found”). How do you run this.
solution
You need to give the path to the executable. A relative path works, so you can just do
./hello
Here, the ./ means “in this directory.
What command do I use to search for all instances of a particular word in a file?

solution

We use grep.

C++ basics#

Suppose I wanted to store the number 0.3 with the most precision possible–what C++ data type would I use?

solution

Since this number has a decimal point, we need a floating point type. We know float and double, but double has more precision (it is double precision, or 64-bit floating point).
How can I compute \(\sqrt{5.0}\)? Is there a particular header file I would need to include?
solution
We could do:
std::sqrt(5.0);
We could also do:
std::pow(5.0, 0.5);
In either case, we need to include the cmath header.
How do I compile a program called power.cpp to make an executable named power?
solution
We do:
g++ -o power power.cpp
When I write std::cout, what is the meaning of the ::?

solution

This is the scope operator. It means “look inside std to find cout”. We call std the namespace.
Consider this code:
```
double x{1.5};
double y = 2.0 + x * 5.0;
```
what is the value of y?

solution

The thing to remember here is that multiplication, *, has higher precedence than addition, +. So this is 2.0 + (1.5 * 5.0) or 9.5.
Consider this code:
```
int x{2};
int y = 2 * x * x / 3;
```
what is the value of y?

solution

All operators here have the same precedence, so we go left to right. Just computing this we would get 8/3, but the result is an int, so the decimal part is discarded, and we get 2.
Consider this code:
```
int x{2};
int y = 2*-x;
```
what is the value of y?

solution

Here, the - is the unary minus or “negation operator”. It simply changes the sign of x, and it has the highest precision here. So this is equivalent to 2 * (-2).
Every C++ program needs a function with what name?

solution

main
Consider the following code:
```
double x{1/3};
std::cout << 3 * x << std::endl;
```
what value is output to the screen?

solution

The key here is that 1/3 is integer math, so this evaluates to 0, and as a result, 0 is output to the screen.
Consider the following code:
```
double x{1.0};
double eps{1.e-30};

bool y = (x + eps) != 1.0
```
what value would you expect y to have?

solution

Compared to 1.0, the number 1.e-30 (or \(10^{-30}\)) is very small, and it is below the roundoff precision, so we would get 1.0 + 1.e-30 = 1.0. (Remember that machine epsilon is ~ 1.e-16).

The != relation operator is not equals, so this is effectively asking if 1.0 is not equal to 1.0, and the answer is false, so y gets set to false or 0.

In-Class Review I

Contents

In-Class Review I#

Bash shell#

C++ basics#