Create a Template Numbers Class

The typedef approach is great because we only have to change one line to change a class from using "int" data to using "float" data. The problem is that we can not easily have a main program that works with BOTH kinds of classes at the same time. This is why templates were added to C$++.$ By using a template class, we can supply the data type we want to use when we declare the objects in the main program.

Start by making a backup copy of the typedef version of "numbers.cpp$"$ so you do not lose this work. Next, edit the class definition to remove the typedef line and add the following "template $"$ just above the class definition. Do NOT put an $"$;$"$ after this line or you will get a million syntax errors.

Next, copy$/$paste this line just before each of your method definitions to tell the compiler that these methods are templated. Then change the "Numbers::" to be "Numbers::" in each of the method headers. When you are finished, your methods should all look like this:

template

int Numbers::getCount$()$

$\{$

return Count;

$\}$

Next, replace the original main program with the following code and compile and run your program. If you look at this code, you will see that we used "Numbers num;" to declare a "num" object that reads and processes the data using "int" data. We used "Numbers num$2$;$"$ to declare the "num$2"$ object which is used to read and process the data using "float" data.

$//$ For templates we MUST compile the numbers.h and numbers.cpp code

$//$ at the same time as main.cpp in order for the compiler to resolve

$//$ what data types you are using in Numbers objects. This is done

$//$ by including "numbers.cpp$"$ below.

$//$ When compiling on the Linux command line use g$++-$Wall main.cpp

$//$ When compiling using an IDE, you may need to rename "numbers.cpp$"$

$//$ to "numbers.cpp$.$txt$"$ and change the include line below accordingly

$//$ If you do not rename the file, it may be compiled twice causing

$//$ a large number of confusing errors.

#include "numbers.cpp$"$

int main$()$

$\{$

string filename;

cout $<<$ "Enter filename:";

cin $>>$ filename;

$//$ Process int numbers

Numbers num;

num.readFile$($filename$)$;

cout $<<$ "min $="<<$ num.findMin$()<<$ endl;

cout $<<$ "max $="<<$ num.findMax$()<<$ endl;

cout $<<$ "mean $="<<$ num.findMean$()<<$ endl;

$//$ Process float numbers

Numbers num$2$;

num$2.$readFile$($filename$)$;

cout $<<$ "min $="<<$ num$2.$findMin$()<<$ endl;

cout $<<$ "max $="<<$ num$2.$findMax$()<<$ endl;

cout $<<$ "mean $="<<$ num$2.$findMean$()<<$ endl;

$\}$

Finally, edit your program and see what happens when you use the following data types $($short$,$ long, double, char, string$)$ when creating Numbers objects. Which work fine? Which have problems?