list-processes and signal-process don't communicate

list-processes and signal-process don't communicate

[Kai Großjohann]

I'd expect that the output of M-x list-processes RET helps me for a
subsequent M-x signal-process RET.  Alas, this is not the case:
list-processes doesn't print the process id and signal-process
doesn't know about process names.

A simple workaround would be to have list-processes show the pid.

kai
-- 
~/.signature is: umop 3p!sdn    (Frank Nobis)

Re: list-processes and signal-process don't communicate

[Kim F. Storm]

grossjoh@lothlorien.cs.uni-dortmund.de (Kai Großjohann) writes:

> I'd expect that the output of M-x list-processes RET helps me for a
> subsequent M-x signal-process RET.  Alas, this is not the case:
> list-processes doesn't print the process id and signal-process
> doesn't know about process names.

Allowing the user to enter the name of a process in addition to
a numeric pid to signal-process would be trivial enhancement to
signal-process:

Index: process.c
===================================================================
RCS file: /cvs/emacs/src/process.c,v
retrieving revision 1.382
diff -c -r1.382 process.c
*** process.c	24 Oct 2002 08:03:41 -0000	1.382
--- process.c	25 Oct 2002 13:15:28 -0000
***************
*** 5391,5404 ****
  }
  
  DEFUN ("signal-process", Fsignal_process, Ssignal_process,
!        2, 2, "nProcess number: \nnSignal code: ",
!        doc: /* Send the process with process id PID the signal with code SIGCODE.
! PID must be an integer.  The process need not be a child of this Emacs.
  SIGCODE may be an integer, or a symbol whose name is a signal name.  */)
!      (pid, sigcode)
!      Lisp_Object pid, sigcode;
  {
!   CHECK_NUMBER (pid);
  
  #define handle_signal(NAME, VALUE)		\
    else if (!strcmp (name, NAME))		\
--- 5391,5430 ----
  }
  
  DEFUN ("signal-process", Fsignal_process, Ssignal_process,
!        2, 2, "sProcess (name or number): \nnSignal code: ",
!        doc: /* Send PROCESS the signal with code SIGCODE.
! PROCESS may also be an integer specifying the process id of the
! process to signal; in this case, the process need not be a child of
! this Emacs.
  SIGCODE may be an integer, or a symbol whose name is a signal name.  */)
!      (process, sigcode)
!      Lisp_Object process, sigcode;
  {
!   Lisp_Object pid;
! 
!   if (STRINGP (process))
!     {
!       Lisp_Object tem;
!       if (tem = Fget_process (process), !NILP (tem))
! 	process = tem;
!       else
! 	process = Fstring_to_number (process, make_number (10));
!     }
!   else
!     process = get_process (process);
! 
!   if (NILP (process))
!     return process;
! 
!   if (INTEGERP (process))
!     pid = process;
!   else
!     {
!       CHECK_PROCESS (process);
!       pid = XPROCESS (process)->pid;
!       if (!INTEGERP (pid) || XINT (pid) <= 0)
! 	error ("Cannot signal process %s", SDATA (XPROCESS (process)->name));
!     }
  
  #define handle_signal(NAME, VALUE)		\
    else if (!strcmp (name, NAME))		\


> 
> A simple workaround would be to have list-processes show the pid.

A more complex approach would be to make the list-processes buffer
obey a mouse-3 click on a process to pop-up menu which allowed you to
terminate them using the mouse.  [and suitable keyboard short-cuts as
well of course].

-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: list-processes and signal-process don't communicate

[Kim F. Storm]

storm@cua.dk (Kim F. Storm) writes:

> grossjoh@lothlorien.cs.uni-dortmund.de (Kai Großjohann) writes:
> 
> > I'd expect that the output of M-x list-processes RET helps me for a
> > subsequent M-x signal-process RET.  Alas, this is not the case:
> > list-processes doesn't print the process id and signal-process
> > doesn't know about process names.
> 
> Allowing the user to enter the name of a process in addition to
> a numeric pid to signal-process would be trivial enhancement to
> signal-process:
> 

Please disregard the patch I just sent; it breaks existing use of
signal-process!  

I'll make a proper patch if there is any interest in adding this.

Re: list-processes and signal-process don't communicate

[Kai Großjohann]

storm@cua.dk (Kim F. Storm) writes:

> Please disregard the patch I just sent; it breaks existing use of
> signal-process!  
>
> I'll make a proper patch if there is any interest in adding this.

Well, I'd be interested, and Richard at least isn't opposed, so if
it won't cause you too much trouble...

kai
-- 
~/.signature is: umop ap!sdn    (Frank Nobis)

Re: list-processes and signal-process don't communicate

[Richard Stallman]

It seems like a useful change--please install it.

Would someone like to add a suitable command to the list-processes buffer?