Path: utzoo!news-server.csri.toronto.edu!cs.utexas.edu!usc!elroy.jpl.nasa.gov!ucla-cs!mara!agsm!iwelch
From: iwelch@agsm.ucla.edu (Ivo Welch)
Newsgroups: comp.sys.next
Subject: gnuplot/gdb/fep (WAS: Re: Fast and/or Nice PLOTS ?)
Message-ID: <822@mara.cognet.ucla.edu>
Date: 15 Mar 91 03:59:48 GMT
Sender: news@mara.cognet.ucla.edu
Distribution: na
Organization: UCLA, Anderson Graduate School Of Management
Lines: 39


Now all we need is a nice NeXTStep frontend to  gnuplot... Any takers?
Which  reminds me: Is   there a  nice  NeXTStep  frontend to gdb?  I'd
certainly shell out money for both of these.

Which reminds  me: is the  john myer's emacs frontend,  (or the dearly
missed NLoad) making any progress on 2.0?


Incidentally, I  fixed   the worst problems    with    fep (which adds
Emacs-style  commands  to arbitrary  programs  invoked  from  a shell;
available at uunet.uu.net/comp.sources.unix).

	(1) replace CTRL(x) definitions to CTRL('x')
	(2) here is the changed function (in fep_set.

set_only_var (name, value)
    char *name, *value;
{
    VAR *vp; void *tempptr=NULL;

    vp = getvp (name, 1, 0);

    /**** this code here is buggy; the initialization calls this with
      "editmode" and "emacs", "emacs" is free-d, and then reassigned.
    if (vp->v_value)
              free (vp->v_value);
    vp->v_value = allocAndCopyThere (value);
    ****************/

    if (vp->v_value) tempptr=vp->v_value;
    vp->v_value = allocAndCopyThere (value);
    if (tempptr) free(tempptr);
}

It still has problems with properly passing  through signals,  such as
CTRL-Z and CTRL-D. Cel a vis.

/ivo welch