Re: HAMT iterator

[Marc Nieper-Wißkirchen]

Am So., 11. Okt. 2020 um 14:04 Uhr schrieb Bruno Haible <bruno@clisp.org>:
>
> Marc Nieper-Wißkirchen wrote:
> > > > /* Return an independent copy of ITER that is initially in the same
> > > >    state.  */
> > > > extern Hamt_iterator *hamt_iterator_copy (Hamt_iterator *iter);
> > >
> > > Then a copy function is not needed, because the user's program can do
> > >
> > >   Hamt_iterator iter_clone = iter;
> >
> > The hamt itself has to be copied (to increase the reference counter).
>
> Then the comment should clarify what "independent" means. I thought it
> means that both iterators (the original one and the copy) can be used
> simultaneously, as long as the HAMT does not change. Do you mean
> something broader?
>   - If someone creates a derivative of the HAMT, the iterator won't
>     be affected, right? ("persistence")

Yes.

>   - If someone makes destructive modifications to the HAMT (through the
>     *_x functions), the iterator will be affected if it has not yet
>     passed the point of modification, right?

No, the iterator shouldn't be affected. (The point of modification is
not well-defined without exposing implementation details.)

> So, what is the scenario where increasing the reference count will make
> a difference?

If you have a language with GC like Lisp or Scheme, say, the hamt may
be GC'd while the iterator is still live.