Replace an element into a specific position of a vector

vec1[i] = vec2[i]

will set the value of vec1[i] to the value of vec2[i]. Nothing is inserted. Your second approach is almost correct. Instead of +i+1 you need just +i

v1.insert(v1.begin()+i, v2[i])

You can do that using at. You can try out the following simple example:

const size_t N = 20;
std::vector<int> vec(N);
try {
    vec.at(N - 1) = 7;
} catch (std::out_of_range ex) {
    std::cout << ex.what() << std::endl;
}
assert(vec.at(N - 1) == 7);

Notice that method at returns an allocator_type::reference, which is that case is a int&. Using at is equivalent to assigning values like vec[i]=....


There is a difference between at and insert as it can be understood with the following example:

const size_t N = 8;
std::vector<int> vec(N);
for (size_t i = 0; i<5; i++){
    vec[i] = i + 1;
}

vec.insert(vec.begin()+2, 10);

If we now print out vec we will get:

1 2 10 3 4 5 0 0 0

If, instead, we did vec.at(2) = 10, or vec[2]=10, we would get

1 2 10 4 5 0 0 0