On Thu, Aug 7, 2008 at 5:25 PM, Roger Pack <roger.pack at
leadmediapartners.com> wrote:
> Should on_resolve_failed call on_close by default? That might be good
> for newbies but also slightly confusing at the same time, since when
> you override it it would no longer call on_close. I''d imagine
with
> proper documentation it wouldn''t cause too much grief.
Not really sure about this one. It might be good not to do it just to
minimize the number of funcalls being made. Also, if resolution fails the
connection was never open in the first place, so I don''t think
it''s
necessarily correct per the method name.
> Also should we add an on_resolve call? Or is that too internal? [It
> would be useful for connection timeouts, but...may not be too
> generically useful].
>
That sounds as if it might be useful. Perhaps it could be used as part of
converting Rev::Sockets to allow creating with .new before the socket is
opened (i.e. you could have a Rev::Socket#open method that on_resolve calls
per default.)
> I''ll probably next work on adding connection timeouts and liveness
> timeouts to rev_em.
>
Cool
--
Tony Arcieri
medioh.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://rubyforge.org/pipermail/rev-talk/attachments/20080811/5fab12e9/attachment.html>