Re: process-attributes patch

[Eli Zaretskii <eliz@gnu.org>]

> From: Miles Bader <miles@gnu.org>
> Cc: Eli Zaretskii <eliz@gnu.org>,  emacs-devel@gnu.org
> Date: Wed, 04 Feb 2009 07:34:43 +0900
> 
> > I generally agree with the change, but I think it comes a bit late.
> > Can't it wait for 23.2?
> 
> The only problem with delaying name changes past a release is that
> user code starts to use old name.

Right, so we should at least change the name of the function.

> The only potential problem I see with this change is compilation
> errors on DOS and W32, but anything there would pretty much be trivial
> to fix for someone who actually has those platforms.  There doesn't
> seem all that much scope for subtle problems.  [Eli found one problem;
> the other thing which may be an issue is using pid_t in the "no
> subprocesses" branch of the ifdef, but that's easy enough to fix.]

Since the "no subprocesses" branch is currently used only by the DOS
port, I can say that there's no problem with pid_t: the DOS port
supports the pid_t type.  ("No subprocesses" does not mean "no
processes".)