associative_vector: evolutionary improvements

From: David Abrahams (abrahams_at_[hidden])
Date: 2000-06-29 19:43:28

John Potter did a lot of work on the associative/indirect container
framework just before he went on vacation. In case anyone's interested, I've
uploaded the latest stuff he sent me to:

http://www.egroups.com/files/boost/associative_vector/associative.zip

Haven't looked through it yet, but mean to get to it "real soon now" ;)
He and I collaborated on general architecture, but he wrote all the new
code.

Jeremy, if you're interested, you might want to see how well this works with
your versions of indirect_iterator.

Here's our last communication:
----- Original Message -----
From: "John E. Potter" <jpotter_at_[hidden]>
To: "David Abrahams" <abrahams_at_[hidden]>
Sent: Friday, June 23, 2000 11:55 AM
Subject: Re: associative_vector: request for formal review

>
>
> On Wed, 21 Jun 2000, David Abrahams wrote:
>
> > I just wanted to express my appreciation for all your work on this, and
to
> > let you know that I'm not ignoring you. I just got a little (!) swamped
but
> > I'll look everything you've sent me over in the next couple of days.
I'll be
> > in touch.
>
> Out of time.  Gone for a month.
>
> Status:
> Vector_set parameterized on Container, nice.
> Associative_vector parameterized on Container, I don't like it but
>   do not have time to undo it.  See what you think.
> Associative_indirect_vector.  Not parameterized on Container.  Needs
>   try/catch in copy_ctor and insert.
> Vector_set_test #ifdefs to control Container.
>
> vector_set<sgi_tree>, vector_set<associative_indirect_vector>,
> vector_set<associative_vector<vector> >,
> vector_set<associative_vector<deque> >,
> vector_set<associative_vector<list> >
> all pass the tests.
>
> Some changes will be needed to remove gnu patches and make usable
> with other compilers.
>
> Will send each of the files in one e-mail.  Do whatever, and I will
> handle any flack next month.
>
> Later,
> John
>