Re: [Sbcl-devel] sb-bsd-sockets and sb-posix fail tests on SunOS

[<ja...@un...>]

Vincent Arkesteijn <vi...@ar...> writes:

> Instead of making sb-bsd-sockets depend on sb-posix, another option
> could be to make sb-bsd-sockets part of sb-posix?
>
> I don't really oversee all the consequences, but everything in
> sb-bsd-sockets seems to fit the scope of sb-posix ("operating system
> calls on a typical Unixlike platform").

I like this idea, but I was kind of under the impression that
sb-bsd-sockets isn't really a one-to-one map to the sockets system,
but is a Lispy interface to socket handling. So sb-posix should have
the raw socket functions and structures that are documented in Unix
manpages, and sb-bsd-sockets should be built on top of that support in
sb-posix. That's kind of how I thought things were headed anyway.

BTW, I care about this because I'm playing around with a telnetd
implementation in SBCL. I'm trying to figure out where to add support
for getaddrinfo(), gettosbyname(), and a bunch of other crufty socket
stuff that the generic BSD telnetd implementation uses.

'james

-- 
James A. Crippen <james at unlambda.com> Lambda Unlimited
61.2204N, -149.8964W                     Recursion 'R' Us
Anchorage, Alaska, USA, Earth            Y = \f.(\x.f(xx))(\x.f(xx))