About Templates

[mike_g]

I just had a go at using templates working off the tutorials here. i get their point and what they do, but there are some things I don't understand yet. For example whats the difference between template <class Something> and template <typename Something>?

Also, in my code I would like to be able to check if a number is greater or equal than the number of bits in my storage type. What would be the best way to do it with this? This seems to work, but I think I read somewhere that sizeof() should only really be used for mallocing:

if(bit >= sizeof(IntType)*8) return false;

Oh and why do I have to put <IntType> between the object name and the function name as in here:

template <class IntType> bool Flag<IntType>::FlipBit(Uint8 bit)

There seems to be no point to it. Anyway, I'll shut up now

[Wraithan]

google template typename class

sizeof could be used for that, assuming you know your target architecture (to make sure it is 8 bits a byte)

last one I would have to look up because I am lame when it comes to templates, and still just learning like you.

[Salem]

8 should be replaced with the CHAR_BIT constant.

[mike_g]

Oh yeah, I'm sure someone told me that before. I knew there was something wrong with it. Cheers.

[brewbuck]

I just had a go at using templates working off the tutorials here. i get their point and what they do, but there are some things I don't understand yet. For example whats the difference between template <class Something> and template <typename Something>?

No difference for most usages.

Also, in my code I would like to be able to check if a number is greater or equal than the number of bits in my storage type. What would be the best way to do it with this?

std::numeric_limits<T> may come in handy.

Oh and why do I have to put <IntType> between the object name and the function name as in here:

template <class IntType> bool Flag<IntType>::FlipBit(Uint8 bit)

There seems to be no point to it. Anyway, I'll shut up now

Don't honestly know the answer to that question, but it has irritated me in the past also. It probably has to do with an ambiguity in the language, but I can't imagine what it might be.

[CornedBee]

The reason for the third is that Flag is a template, thus you must specify what you're implementing. Just because there's one template argument there doesn't mean that's exactly the template argument you'd use for the class. What if you're implementing a template member? What if you're partially specializing?