When is the C `typename` Keyword Necessary?

When is the "typename" keyword necessary?

In C , the "typename" keyword is necessary when referring to a nested name that is a dependent name, i.e. nested inside a template instance with unknown parameter.

A dependent name is a name that cannot be fully resolved at compile time because it depends on a template parameter. For example, in the following code:

template<class K>
class C {
    struct P {};
    vector<P> vec;
    void f();
};

template<class K> void C<K>::f() {
    typename vector<P>::iterator p = vec.begin();
}

The name vector

::iterator is a dependent name because it depends on the template parameter K. At compile time, the compiler does not know the value of K, so it cannot fully resolve the name vector

::iterator.

Tthe "typename" keyword is used to tell the compiler that the name vector

::iterator is a type name, even though it cannot be fully resolved at compile time. This is necessary because the compiler needs to know the type of the variable p in order to generate correct code.

There are other cases where "typename" must be specified, such as when referring to a nested type that is declared inside a template class. For example, in the following code:

template<class T>
struct A {
    typedef typename T::type my_type;
};

The name typename T::type is a nested type name that is declared inside the template class A. At compile time, the compiler does not know the type of T, so it cannot fully resolve the name typename T::type.

The "typename" keyword is used to tell the compiler that the name typename T::type is a type name, even though it cannot be fully resolved at compile time. This is necessary because the compiler needs to know the type of the variable my_type in order to generate correct code.

The above is the detailed content of When is the C `typename` Keyword Necessary?. For more information, please follow other related articles on the PHP Chinese website!