Exercise 3 (Week 4) File handling

SCIT-EIS-UOW
CSCI251/CSCI851 Advanced Programming

Laboratory Exercise 3 (Week 4)
1 Warm–up exercises
1. File handling exercise: Warmup-file.cpp. This isn’t debugging and the program should compile
using CC on Banshee without problems.
(a) How are the input elements separated?
(b) Enter a few valid lines of data, end the program, and look at the content of the data file. You
can use cat to view the file.
(c) Run the program again, entering some more data before terminating the program. What has
happened to the data file?
(d) Run the following:
$ mv Students.txt Data.txt This is renaming the file.
$ ./a.out < Data.txt This is directing input from a file.
(e) Modify Warmup-file.cpp to open the file in append mode instead of the default write mode.
Careful with the scope.
(f) Re-run the program. What happens to the data file now?
2. Vector handling : Warmup-vector.cpp is similar to the Warmup-file.cpp but goes through the
syntax for adding elements to vectors.
3. Two files to fix: Debug-B1.cpp and Debug-B2.cpp. How do these differ? Fix them, compile them
and check how the executables differ.
4. Memory leak:
int *p;
p = new int(5);
Use bcheck to confirm some memory is leaking. Fix it and use bcheck again to confirm it’s fixed.
You can run bcheck as follows:
$ bcheck ./a.out
5. Demonstrate the use of cerr to report errors. You can direct the error output to somewhere using
this type of instruction from the command line.
$ ./a.out 2> Errors.txt
Standard error will be redirected to the file Errors.txt.
1
6. What does the following code do, with C++11 compilation?
int array[10]={3, 5, 7, 10, 3, 5, 4, 4, 9, 3};
for (auto &element : array)
cout << element * element << endl;
What about using this?
int *array = new int[10]{3, 5, 7, 10, 3, 5, 4, 4, 9, 3};
You should be able to write an ordinary for loop to make sure the values are in there, but the for
from above won’t work because it’s dynamic memory. The error message indicate begin and end
functions used to work out the range don’t work! More on those later in the subject.
2 Task Two: File handling
1. Make a copy of the file /usr/dict/words in your directory. Write a program to read words from
the words file, and output them together with the line number of each word. Include appropriate
checks on the opening of the files. Think about the order you do the checking in.
2. Test each of the open modes for files. See what effect each has on a text file you generate.
3. Write a program to write some text to a file, and then read the file back in binary mode.
4. Write a program to read in a file full of digits and produce a digit distribution. You can use
the file digits.txt as sample data. Output the digit distribution in a sensible format to a file
Distribution.txt.
5. * How could you find the size of the output buffer?
6. * How about the size of the input buffer?
3 Task Three: Investigating random
1. The file Old-Random.cpp uses the old style randomness.
(a) Run it to see what the output is like.
(b) Run it several times immediately after each other. Using the following might help:
$ ./a.out ;./a.out
What is happening? Why?
2. Copy Old-Random.cpp to a new file Old-Random-Count.cpp and modify this new file so the program
produces N integers in the range 0 to 9 inclusive, so effectively digits, with N being a parameter
given on the command line. Set up an integer array to count how many of each digit occurs. Output
a sensible description of the distribution.
3. Write a program in New-Random.cpp that uses the C++11 code in the lecture notes to that uses the
randomness engine to produce integers in the range 0 to 9 inclusive from a uniform distribution. Use
similar code to the above to output the distribution. Again the number of random digits produced
should be based on a command line argument.
4. Write a new program in Random-normal.cpp that is similar to the above but runs as
2
./Random-normal N mean stdev
where N is the number of digits to be generated, mean is the mean of the normal distribution, and
stdev is the standard deviation of the distribution. The output for this version should be rounded
to the nearest integer.
5. * Run your Random-normal.cpp program and produce test data for the three parameters being
10000 1000 100. Count the different number of outcomes in the range 500 to 1500.
6. * Mersenne Twister: This is a family of pseudo–random number generators, the best known member
being Mersenne Twister 19937.
std::random_device randev;
std::mt19937 mt_eng(randev());
4 Task Four: Basic struct construction
Carry out the debugging exercise first, Debug-struct.cpp.
Create a piece of code Dog.cpp. This code should contain a struct Dog with the following properties:
• Data fields for name, breed and age.
• A constant static field for the license fee, which is $17.55.
• Functions to set and display the data. You are not to use explicit constructors.
• A main() function that demonstrates the struct operates correctly.
Dog: Flash is a Spotty thing.
The dog’s age is 4
License fee: $17.55
5 * Task Five: Memory ...
How much memory can you allocate on Banshee using new?
c Luke McAven, SCIT-EIS-UOW, 2019.
3