How to implement the header of a function such that the following code works?

Question

vector<int> v1{4, 2, 1, 6, 3, -4};
assert(fct<int>(v1) == 6);
vector<int> v2;
try {
        fct<int>(v2);
        assert(false);
}
catch (exception& exc) {
assert(true);
}
vector<double> v3{2, 10.5, 6.33, -100, 9, 1.212};
assert(fct<double>(v3) == 10.5);

vector<string> v4{"y", "q", "a", "m"};
assert(fct<string>(v4) == "y");

I know that the fct function should return the maximum of the vector but I can't wrap my head on how the header of the function should look like.

If this is not a homework, you should be using [`std::max_element`](https://en.cppreference.com/w/cpp/algorithm/max_element). — n. m. could be an AI, Jun 21 '21 at 08:45
there is no `fct` that can make `assert(false)` pass and `assert` does not throw an exception that you can `catch` — 463035818_is_not_an_ai, Jun 21 '21 at 09:10

Stefan Riedel · Accepted Answer · 2021-06-21T12:23:11.450

3

I guess you meant the function declaration (not "header"), so that's what it should look like:

template<typename T>
T fct(const std::vector<T>& v);

PS:

You should not name that function "fct", but give it an appropriate name, like "max_element" or something (or use the function std::max_element provided by the standard lib)
Don't use using namespace std; see: why is 'using namespace std;' considered bad practice
~~If you're not bound to some older C++ standard (or compiler),~~ you can omit the template parameter when calling your function: assert(fct(v1) == 6);

edited Jun 21 '21 at 12:23

answered Jun 21 '21 at 08:43

Stefan Riedel

796
4
15

I don't think there was a C++ standard without template parameter deduction from function parameters. Are you refering to C++17s CTAD? `fct(v1)` isn't that. – 463035818_is_not_an_ai Jun 21 '21 at 09:12
Right, my bad. That was just class template argument deduction. Edited the Answer. – Stefan Riedel Jun 21 '21 at 12:23

How to implement the header of a function such that the following code works?

1 Answers1

PS: