Re: [boost] asio formal review begins

From: Arkadiy Vertleyb (vertleyb_at_[hidden])
Date: 2005-12-14 15:10:30

"Stefan Seefeld" <seefeld_at_[hidden]> wrote

> Arkadiy Vertleyb wrote:
>
> > I would expect both asio and a stream API to be layered on top of
low-level
> > socket API.  This socket API should wrap the platform dependency, but
> > otherwise provide pretty much the same capabilities as current C socket
APIs
> > provide, with minimum overhead.
>
> Yes. Though I'm not sure a thin wrapper around the platform API adds much
> value at all. Too big are the differences. And besides, why should people
> really care whether the thing is a socket, as long as the semantics are
> correct ?

>From the usage point of view there might be little difference, as long as it
works correctly, and doesn't add a lot of performance overhead.

But from the design point of view, I think it's important.  And if the
socket class is used to perform synchronous IO, I don't think the user
should have to create the demuxer object and carry it around.  Also, IMO,
something that is presented as a "Socket" should not hide things that are
conceptually related to the socket usage, rather than the socket itself.
These things should be located elsewhere, so that they become relevant only
if used.

Regards,
Arkadiy