matplotlib-users Mailing List for matplotlib

Hi
1.)
Hmmm the approach described below cycles colors for
	
	rcParams['lines.marker'] = 'o'
but for
	rcParams['lines.marker'] = '+'
I get black markers all the time (no cycling).
2.)
I'm wondering if there is an 'elegant' way to force color cycling for
things like
	for i in range(...):
		plot(...,'+')
	for j in range(...):
		plot(...,'x')
or do I always have to set
	rcParams['lines.marker'] = <makrer>
cheers,
steve
John Hunter wrote:
>>>>>>"Steve" == Steve Schmerler <el...@gm...> writes:
> 
> 
> Steve> Hi Well if you do lineplots
> 
> Steve> 	# plot some lines x = [1,2,3] for i in range(...):
> Steve> plot(x)
> 
> Steve> mpl changes the color of each line which doesn't happen in
> Steve> the case of marker plots.
> 
> Actually something different is going on, but I had to grok through
> matplotlib.axes._process_plot_var_args to figure it out.
> 
> The default color cycling happens when there is no string format
> applied, and is independent of markers and lines. For example, the
> following does not cycle either
> 
> for i in range(4):
> plot(rand(5), rand(5), '-')
> 
> 
> because a format string is applied. 
> 
> You can make markers cycle too w/o a form string by changing the rc
> params so that the default makrer is not 'None'
> 
> rcParams['lines.marker'] = 'o'
> rcParams['lines.linestyle'] = 'None'
> for i in range(4):
> plot(rand(5), rand(5))
> 
> 
> Whether or not this is ideal behavior is debatable. But it is
> probably good enough since it is easy enough to force plot to act like
> you want by explicitly passing args, as you did. I think Niklas'
> suggestion of explicitly passing the kwargs for marker, linestyle,
> color, markerfacecolor and so on is a better approach than
> constructing arcane format strings. It is more readable and more
> flexible, because format strings limit you to a small set of colors
> whereas the kwargs approach supports arbitrary color arguments.
> 
> JDH
> 
>