Re: List container strawman

[Darren New <dnew@san.rr.com>]

> > If you really want it to be foolproof,
> Yes please.

Well, yes. However, the performance overhead can be significant. There
are ways of reducing it (see, for example, some of the fields in the
private structures in my offering) but to be 100% bullet-proof, you'd at
least need to have O(N) access times. Given user-programmable and
multiple storage pools, this would be harder than in Pascal. The trick
is to make it so you can turn it on and off at runtime, or for only a
particular type of list (i.e., particular instantiations) so you can
narrow down the problem.

Actually, I never really thought of it with user-definable storage
pools. There may be a much more efficient way of working it.

A couple other points:
1) Does Ada even ensure there's only one default storage pool? I would
think that perhaps the storage pool for Integers might be different than
the storage pool for big records.... perhaps a different storage pool
for each size item being stored or some such? Hence, I'm not surprised
that it's difficult to find out what the "default" storage pool is. Or
am I wrong in this?

2) Perhaps using terms from Unbounded_String might be wise? I.e., stuff
that treats an unbounded string as an array of elements that happen to
be characters rather than relying on the fact that they're characters.
Ie., calling the routines Element(), Replace_Element(), Slice(),
Overwrite(), Delete(), "*", etc?  This isn't what I think of as the best
naming convention, but it's at least going to be familiar to other Ada
programmers.

-- 
Darren New 
San Diego, CA, USA (PST). Cryptokeys on demand.
     Sore feet from standing in line at airport
                 security checkpoints: Jet Leg.