Re: GIOChannel diff committed
- From: Ron Steinke <rsteinke w-link net>
- To: hans breuer org, rsteinke w-link net
- Cc: rsteinke w-link net, otaylor redhat com, gtk-devel-list gnome org
- Subject: Re: GIOChannel diff committed
- Date: Sat, 21 Jul 2001 13:09:32 -0700
> From: Hans Breuer <hans breuer org>
>
> Should I change the is_*able flags than, if the underlying system
> call gives a change to these flags ? (Appears to be a little less
> obvious to me).
The only functions I know which could require a change of those
three flags (at least under unix) are close() and shutdown()
(partially close socket connection). The change to
the flags in io_close() is handled in the generic wrapper
functions. At this point, we're not dealing with the possibility
that the user may close the channel with a non-glib function.
This suggests that the is_*able flags should be accessable
in the API, and we need to add a stern warning to the documentation
about having to change those flags when closing a channel without
using the API.
The effect of this is that, other than when the channel is first
created, the backend should never need to set these flags.
Ron
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]