Re: [gtkmm] A const_iterator for TreeModel?

[Robert Zeh <rzeh efs-us com>]

murrayc murrayc com wrote:

> > Take a look at Item 26 in Scott Meyer's Effective STL.  He says that
> > there is an implicit conversion from iterator to const_iterator
> > --- but not the other way around!.
> >    
>
> Interesting. I had tried that with std::list_iterator and const_iterator,
> without success. Could someone else try as well please.
>  
What exactly did you try?    This worked for me, with GNU 3.3.2 and
SunPro:

 std::list<int> l;
 std::list<int>::iterator bar(l.begin());
 std::list<int>::const_iterator foo(bar);
 ///std::list<int>::iterator bar2(foo);   /// Fails --- can't go 
directly from const_iterator to iterator.

> > You have to use the base() method on
> > the reverse iterators to get to iterators.
> >    
>
> Thanks. I didn't know that. Is that something that people are used to
> using, or is it an obscure part of the implementation?
>  
Base() isn't terribly obscure; it's part of the standard, and mentioned 
in Stroustrup.

> > So, you should be able to write a definitive const_iterator method,
> > and then have the other three call the const_iterator method
> > with the appropriate implicit or explicit conversion.
> >    
>
> If there is _an_ implicit conversion then we should not need the other
> method overloads.
>
> _______________________________________________
>  
You're right.  You should be able to just write:

foo(const_iterator)

And call it with

foo(iterator)

Of course, this assumes that everything you are doing now with an 
iterator can
also be done with a const_iterator.

However, you'll still need to have another function definition for the 
reverse_iterator's.
There are some corner cases in converting between reverse iterators and 
forward
iterators.

Robert Zeh