Christian, Christian Schladetsch wrote:
This is precisely what my boost::monotonic proposal addresses. It gives you, the user of the containers, complete control over where the memory comes from,
I didn't follow his proposal closely, but I believe that is exactly what Thorsten Ottosen proposed with his auto_buffer proposal, no? AFAIK he has offered a fairly polished implementation, which is now awaiting review.
and how much there is of it.
In a std::vector, isn't that what vector::reserve is for (run-time reservation)? In a boost::auto_buffer, isn't that what StackBufferPolicy is for (compile-time reservation)? As has been asked before by Frank Mori Hess in this thread, and as I don't believe you have answered yet, in what respects is your proposal better/different from Thorsten's auto_buffer? Thanks, François