On Fri, Jan 11, 2002 at 01:29:33PM +0100, Sander Striker wrote:
> > But that's NOT really the point is it? The point is that not all the
> > platforms can truly support this functionality and adding it just so we have
> > it on some platforms and everyone else has to return ENOTIMPL seems to be
> > crazy and against the very reason for APR.
>
> I agree. We need a portable ipc mechanism.
Hehe. I believe Aaron and Will have some ideas for this (they call
it spipe, IIRC). I'll let them post their ideas rather than me try
to mangle what I've been told. I think they've been busy with the
shmem stuff that this has gotten pushed back a bit.
AIUI, their idea is completely separate from Unix domain sockets,
but is definitely the core IPC piece that can be implemented
portably.
> It belongs in APR, because otherwise you have to do runtime checks in apr-util
> to see if APR_ENOTIMPL is returned for unix domain sockets and then try the
> thing we have as a fallback for a different platform (yuck).
Yup.
I think the unix domain socket change is very minimal. The only
thing we'd really need to add is logic to handle the path setting
portably (i.e. apr_sockaddr_path_{get|set}). But, it isn't true
portability in that it'll run on every platform (apr_sendfile
anyone?). -- justin
