Re: [Gnuplot-py-users] quick and dirty catch-all plot function

[Michael H. <mh...@al...>]

On 02/21/2011 09:51 PM, John E Field wrote:
> I installed Gnuplot.py and I like it a lot. How about a really quick and 
> easy routine to just plot a function or several functions. It seems like 
> I am plotting things so much that I shouldn't have to write several 
> lines of code to make it happen.  Perhaps it is already in there and I 
> just don't know about it. Something sort of like this :

I'm not such a fan of magical functions that try to guess what you mean.
 The problem is that everybody has a different opinion about what magic
they prefer.  I prefer flexible but straightforward functions that can
be *used* to make magical functions.  So I wouldn't advocate putting a
function like this in Gnuplot.py, though it is a fine example.

> gplot(x,y) <- plots y vs x
> gplot(x,y,with_='lines',title='my title') <- adds title and plots with lines
> gplot(x,y,z) <- plots y and z versus x
> gplot([1,2],sin) <- plots sin(x) from 1.0 to 2.0
> 
> and, the code might look something like this :
> 
> def gplot(*xy,**args) :
>    """ routine for plotting a function using gnuplot directly from python
>        usage: pass arguments in args. for with lines, do with_='lines'
>        title = 'my title'
>        xlabel= 'my xlabel',
>        etc.
>        if the persist option is false, then the plot will only last 
> until the returned object is destroyed
>        otherwise, the plot is forever.
>    """
>    try :
>      if args.has_key('persist') :
>        g = Gnuplot.Gnuplot(persist=args['persist']);
>        del args['persist'];
>      else : g = Gnuplot.Gnuplot(persist=True);
>    except :
>      from Gnuplot import Gnuplot,Data
>      if args.has_key('persist') :
>        g = Gnuplot(persist=args['persist']);
>        del args['persist'];
>      else : g = Gnuplot(persist=True);
>    x = xy[0];
>    if len(x) == 2 :              # fill up the range
>      x = [x[0]+(x[1]-x[0])*0.005*i for i in range(201)];
>    d = [];       # make an argument list for each plot function
>    for yi in xy[1:] :
>      if not (yi.__class__ is list or yi.__class__ is tuple) : yi = 
> map(yi,x);
>      if args.has_key('with_') :
>        di = Data(x,yi,with_=args['with_']);
>      else : di = Data(x,yi);
>      d.append(di);
>    if args.has_key('with_') : del args['with_'];
>    return g.plot(*d,**args);

BTW, there are some things in your function that can be simplified
(untested code):

def gplot(*xy,**args):
    """..."""
    try:
      Gnuplot, Data
    except NameError:
      from Gnuplot import Gnuplot, Data

    g = Gnuplot(persist=args.pop('persist', True));

    x = xy[0]
    if len(x) == 2:
      # fill up the range
      x = [x[0]+(x[1]-x[0])*0.005*i for i in range(201)]
    d = []       # make an argument list for each plot function
    data_opts = {}
    if 'with_' in args:
      data_opts['with_'] = args.pop('with_')
    for yi in xy[1:]:
      if not isinstance(yi, (list, tuple)):
        yi = map(yi,x)
      d.append(Data(x, yi, **data_args))
    return g.plot(*d, **args)

Michael

-- 
Michael Haggerty
mh...@al...
http://softwareswirl.blogspot.com/