usage of keyword template / standard

[Alexander Stippler]

Hi,

I'm not quite sure if I have to use the keyword template in a situation,
where my compiler enforces it, though I think it is not neccessary. My
question is: Who is wrong with respect to the standard - the compiler or I?

template <typename T>
struct A
{
template <typename C>
bool
test();

static size_t value = sizeof(A<T>::test<T>());
};

Do I need to qualify test by 'template'. IMO no.
It would be different in a context like
sizeof(T::template test<T>())
Am I wrong?

regards,
alex

[Attila Feher]

Alexander Stippler wrote:

Hi,

I'm not quite sure if I have to use the keyword template in a
situation, where my compiler enforces it, though I think it is not
neccessary. My question is: Who is wrong with respect to the standard
- the compiler or I?

template <typename T>
struct A
{
template <typename C>
bool
test();

static size_t value = sizeof(A<T>::test<T>());
};

Do I need to qualify test by 'template'. IMO no.
It would be different in a context like
sizeof(T::template test<T>())
Am I wrong?

You are. A<T> is also dependent on T.

--
Attila aka WW

[Attila Feher]

Attila Feher wrote:
[SNIP]

template <typename T>
struct A
{
template <typename C>
bool
test();

static size_t value = sizeof(A<T>::test<T>());
};

Do I need to qualify test by 'template'. IMO no.
It would be different in a context like
sizeof(T::template test<T>())
Am I wrong?

You are. A<T> is also dependent on T.

From the standard, where it says what is dependent:

— a templateid in which either the template name is a template parameter or
any of the template arguments is a dependent type or an expression that is
typedependent
or valuedependent.

Clearly A is a template ID and the template argument is clearly a dependent
type.

--
Attila aka WW