Stem update, rough guess.

[John W. Eaton]

On  6-Apr-2007, Daniel J Sebald wrote:

| Attached is an update to the stem() plot.  Note that one problem with
| the current implementation is
| 
| stem([1:30],'o')
| 
| doesn't properly initialize options.  The reason is that 
__default_plot_options__ looks like the following:
| 
| function options = __default_plot_options__ ()
| 
|   options.key = "";
|   options.color = [];
|   options.linestyle = "-";
|   options.marker = "none";
| 
| endfunction
| 
| and the color of [] fails when it is passed into plot().  I've fixed
| that here by only assigning the color if it is not [] from the
| option settings.  (Same with marker of "none".)  It may the way you
| had in mind John, so change if you like.

Checking that the color is not empty is the thing to do.

| Also, I don't really follow the hh, hhh, hhhh handle stuff.  My way
| of thinking is that plot() should now assign that correctly if
| everything is done with one plot command.  So I've put comment
| characters in front of all those lines and someone else can evaluate
| if that is really needed.

I don't know what that is all about either.  We should probably be
setting the markerfacecolor property if dofill is true, but that
property is not handled in the graphics.cc file or by __go_draw_axes__
yet.  I have no idea what the intent was for plotting an additional
line using the axis limits.

Note that the way things are now, we return a list of grahpics
handles, one for each line in the plot.  But Matlab only returns one
handle, and it is some kind of group object that we don't handle yet.
So doing this right will require some more work (for one, figuring out
exactly how and when we should be creating these group objects).

jwe