Overlapping functionality: IFTI, templates, is-expressions

[Jason House]

Russell Lewis Wrote:

> Jason House wrote:
> > #2 (SFINAE) I know from past discussions that C++ programmers hold this very near and
> > dear to their heart.  Personally, it seems like a back door to long
> > compilation times.  I don't have any problem with enforcing criteria for
> > templates to be defined up front, but I'm certain others don't agree with
> > me.  Does the loss of SFINAE make templates too close to generics?  I know
> > C# will refuse to compile code if the up-front criteria for a generic type
> > isn't specific enough.
> 
> FYI: SFINAE is "Substitution Failure Is Not An Error".  See Wikipedia.


Sorry, I should have defined it in my post.  That's what I get for thinking about too many things at once.




> I thought about this some more, wondering why C++ programmers would have 
> come to this conclusion.  I now think it likely that they did this 
> specifically because they didn't have static if() and is-expressions. 
> If you don't have those two tools, then the only thing you can do is to 
> run ahead into something, and hope it works...but if it doesn't work, 
> you provide a more general alternative.


Sounds right.


> Is it possible that SFINAE was a hack to get around C++'s lack of 
> compile-time features?  A hack that we can now dispense of?

I don't know if it's that simple.  What if I want a specialization where T extends from two interfaces?  I think the current syntax can handle simple stuff, but I don't know if it can go beyond that.  Maybe some would argue the more complex cases should be done with static if anyway...

Personally, I wouldn't mind some kind of optional in clause for a templated definition that can do a few static asserts to force more complex conditions...  I don't know what would be the best syntax for that.