COMP151 Lab 12
template & operator overloading

Objective:

To learn template and operator overloading in C++

Files:

Lab work:

What is a template?

Templates can generate particular instances of a class/function varying by type. We parameterize all or some types in a function/class.

template syntax

// template function
template<typename T> T min(const T& t1, const T& t2) {
  return t1 < t2? t1: t2;
};

// template class
template<typename T> class Array {
public:
  // ...
  T& at(int i);
};

// instantiate template function implicitly
int i = min(10, 20);

// instantiate template function explicitly
double d = min<double>(4, 5.5);

// instantiate template class
Array<int> ai;

//...

Template example

In templatefunc.cpp, min is a template function. Try running it to see its output. Note the use of min<double>(5, 7.50) instead of just min(5, 7.50). Why do explicitly instantiate the function template with a <double>.?What would have happened if we had written min(5, 7.50)?

Now remove the "/*" and "*/" in the main program, and try to recompile the program. The program will not compile because min is called with 2 Dummy objects as parameters. What is the problem? How can you fix it?

What is operator overloading?

Customization of C++ operators. You can overload most of the C++ operators. But you CANNOT invent your own operator!

The following are nonoverloadable operators

:: .* . ?: sizeof

Operator overloading syntax

// global function
ostream& operator<<(ostream& os, const Dummy& dummy) {
  // ...
};

For an example of overloading a  non-global operator, see the overloaded operator += of evector in evector.cpp.

Assessment:

evector is a template class that is publically derived from the template class vector in STL. It inherits all of the vector member functions including the operator[] from vector so that you can access each individual element by [], e.g. ev[5] returns the 6th element.
+= is an overloaded member function. It adds the second parameter to each element in the evector
<< is an overloaded global function. It prints all of the elements in the evector separated by spaces.

To complete evector.cpp, you need to:

Fill in your name, student id and ITSC/CSD account id at the beginning of the evector.cpp as comment
Overload the "-=" operator. This is similar to "+=", except that it subtracts the second parameter from each element in the evector. You may assume that operator"-" is defined on objects of class T.
Overload "*" operator to do "dot product" of 2 evectors. If the sizes of 2 evectors are the same, it should create a new evector such that the new i-th element is equal to the product of the i-th elements of the 2 evectors. If the sizes of the 2 evectors are different, it returns an empty evector. Note that this is a member function You may assume that operator "*" is defined on objects of class T.
Overload the "+" operator to do addition of 2 evectors. If the sizes of the 2 evectors are the same, it should create a new evector such that the new-ith element is equal to the sum of the i-th elements of 2 evectors. If the sizes of the 2 evectors are different, it returns an empty evector. Note that this is a member function. You may assume that operator "+" is defined on objects of class T.
Remove the "/*" and "*/" in the main program
compile the program named as: evector
If your program works properly, it should give the following output

Hints:

the function prototype to overload * is "evector<T> evector<T>::operator*(const evector<T>&)"
the function prototype to overload + is "evector<T> evector<T>::operator+(const evector<T>&)"

Submission:

Put your file into ~/comp151/lab12
The system will collect your files at mid-night of the lab day

This page is created by Chan Ka Ki.

COMP151 Lab 12 template & operator overloading