[comp.emacs] jove and iprocs

windley@cheetah.ucdavis.edu (Phil Windley) (10/12/88)

I'm running Jove on an IBM RT running AIX (SysV with BSD extensions) 2.2.
AIX 2.2 has sockets, ptys, etc, but no job control.  The version of Jove I
have says that you must have job control to use iprocs.  Now obviously,
this isn't true in general, just for Jove.  Has anyone suceeded in
divorcing these two things in Jove such that one can use interactive
processes with OS's that don't have job control?  I'd be interested in the
patches. 

--
Phil Windley                          |  windley@iris.ucdavis.edu
Division of Computer Science          |  ucbvax!ucdavis!iris!windley
College of Engineering                |  (916) 752-7324 (or 3168)
University of California, Davis       |  Davis, CA 95616

jpayne%breakpoint@Sun.COM (Jonathan Payne) (10/13/88)

In article <WINDLEY.88Oct11153446@cheetah.ucdavis.edu.cheetah.ucdavis.edu> windley@cheetah.ucdavis.edu (Phil Windley) writes:
>
>I'm running Jove on an IBM RT running AIX (SysV with BSD extensions) 2.2.
>AIX 2.2 has sockets, ptys, etc, but no job control.  The version of Jove I
>have says that you must have job control to use iprocs.  Now obviously,
>this isn't true in general, just for Jove.  Has anyone suceeded in
>divorcing these two things in Jove such that one can use interactive
>processes with OS's that don't have job control?  I'd be interested in the
>patches. 
>
>--
>Phil Windley                          |  windley@iris.ucdavis.edu

Well, hi.  I have made it so that it is not necessary to have job control
for iprocs to work anymore.  In the event that you are not 4.2 or greater,
JOVE uses pipes to do the iprocs.  It doesn't do it in the yucky way that
it always used to, using signals back and forth, so that's all gone.  I
actually haven't tested this on nonbsd systems so I am not sure how the
waiting for processes to die actually works, but that shouldn't be too much
of a problem.

But I think that if you have sockets, ptys and SELECT, then you will be
able to get by will little or no changes to make this work.  The real
division should take place at the select vs. no select point, since that
is the source of the problem.