Boost logo

Boost :

From: Daryle Walker (darylew_at_[hidden])
Date: 2004-11-18 06:18:26


On 11/7/04 2:54 PM, "Jonathan Turkanis" <technews_at_[hidden]> wrote:

> I can see from the most recent review schedule that the More IO library has
> been rejected, but I have not seen a review result posted. I'm interested to
> know the rationale for the rejection and future plans for More IO, since I
> would like to add the Iostreams library to CVS as soon as possible and several
> of the open issues depend on the status of More IO.

I haven't heard anything about this either until I checked the schedule
after reading this message. I guess it was "pocket" rejected, just like the
review in 2003.

Tom did send me a private message talking about a later (mini-) review after
I tweak the library over the last set of suggestions. I'm still mulling
over what to do next. So don't let me stop you from integrating your code.
What are the specific issues?

> For example, during the review of both libraries there was definite support
> for some sort of null_sink/null_buf. I'd like to know if I should go ahead and
> add these classes to the iostreams library, or whether I should wait for
> Daryle to produce a revised version of More IO.

Nothing prevents both libraries, in some form, from co-existing. Especially
for different users that disagree over the "simpler but completely-new
framework" versus "an extension of the Standard framework" controversy.

> Certainly I can put off some of there decisions until later, but I'd rather
> take a few days and prepare a revision of the Iostreams library all at once.

I guess that your library will be inserted in time for 1.33? Since we don't
have a timetable for that release (yet, AFAIK), a third version of More-I/O
may be ready by the same time (if approved).

-- 
Daryle Walker
Mac, Internet, and Video Game Junkie
darylew AT hotmail DOT com

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk