|
Boost : |
From: davlet_panech (davlet_panech_at_[hidden])
Date: 2002-03-01 09:22:54
--- In boost_at_y..., jk_at_s... wrote:
> 25 Feb 2002 04:54:20 +0300 mfdylan wrote:
> >Error handling almost has to be done with exceptions - and note
that
> >basically all socket errors except EWOULDBLOCK mean the socket is
no
> >longer usuable anyway.
>
> Not exactly. There is also EINTR condition, and the socket is still
> usable after that; and there are ENOBUFS, ENOMEM, EMSGSIZE for
sending,
> which are restartable too - when there is memory shortage, all you
need
> to do is just repeat after some reasonable delay. Furthermore, after
> EPIPE the socket is unusable but it is hardly can be considered as
> runtime-error.
Yes, except that at least some of these conditions do not apply to
non-unices (e.g., there is no EPIPE or EINTR on windows).
D.P.
>
> --
> jk
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk