Re: [boost] Proposal: Monotonic Containers

Subject: Re: [boost] Proposal: Monotonic Containers
From: Scott McMurray (me22.ca+boost_at_[hidden])
Date: 2009-06-09 22:41:58

2009/6/9 Cory Nelson <phrosty_at_[hidden]>:
>
> I would want to see it taken a step further though. Â Overhead can be a
> big deal for large objects, so re-implementing string and vector
> completely (so that you don't have the pointer/capacity specified
> twice, in container & allocator) would be a welcome change.
>

An implementation could already define new/delete and std::allocator
in such a way that the allocation block size wasn't stored in the
allocation system.

Are there any that do so?

(And -- if we could ignore backwards compatability -- would it be
worth requiring delete[n] so that the same could be done for
new[]/delete[]?  My gut feeling is that programmers have that
information around at delete[]-time anyways...)

~ Scott