Folks,
Could anyone help me understand how do you decide whether to use templates or overloaded function? I was asked this question in an interview and couldn't answer. This is the scenario:
Given an algorithm, how do you decide whether to use a template method or overloaded function?
Thanks for your insights!
Someone may give a better explanation but I would say that, using a template would mean you only have to implement the function once rather than many times for different types in the case of overloaded functions.
I'd say, if the new functions take exactly the same number of parameters and do exactly the same things inside the function (other than using different data types), then you should use a template function. Otherwise you'll be copying & pasting functions and only changing data types.
Technically, a template function is a means of creating a set of overloaded functions.
As a rule of thumb, I will employ template functions if I need an overloaded function that works for many argument types and, if I was to implement every possible version of that function by hand, the body for most of those versions would be essentially the same.