Hi Tom,

>  1. When I naively combine LineInspect used in two_plots.py with the
>  range_selection behavior in zoomed_plot/zoom_plot.py.  The
>  LineInspector doesn't show up.  From the way the mouse moves, it looks
>  like it is doing something but I don't see the lines.
>  2. Something is strange with the interaction of backbuffering and the
>  right y axis.  If I set it to right and resize the window horizontally
>  the y axis always shows up in the same offset from the left of the
>  window which puts it in the middle of the graph, if you increase the
>  size of the window, and off the side of the window, if you decrease
>  the size of the window.  Strangely it works, however, if you maximize
>  the window.

I'll look at these two and get back to you.

>  3. I have no idea from the examples how to get multiple lines on
>  multiple plots.  I.e. I want two vertically stacked plots that each
>  have multiple lines.  I see examples for each independently but I
>  don't understand how to combine them.

Here are some general guidelines, let me know if they help:

1. All containers are nestable.  So you can three  
OverlayPlotContainers in a VPlotContainer and they will be stacked  
vertically.  Inside each of these OverlayPlotContainers, you can add  
an arbitrary number of overlapping LinePlots.

2. If all of the plots in a given overlapping area share the same X  
and Y ranges, then it's more convenient to use a Plot object instead  
of an OverlayPlotContainer.  Most (or all) of the examples in  
examples/basic/ use Plot.  For instance, in line_plot1.py, there are  
two Plots in an HPlotContainer.  The first one, plot1, has multiple  
line plots on it.  (You can think of the Plot object as sort of like  
MPL's object-oriented plot interface.  Thus, you can call its .plot()  
method with a multiple Y values for a single X value, toggle axes on  
and off, etc.)

3. If you can't or don't want to use the Plot object, then you can  
create LinePlots (perhaps using the helper functions from  
plot_factory.py) and place them inside OverlayPlotContainers, then  
place the OverlayPlotContainers inside other plot containers.

>  4. Finally I want to get it to format as dates.  It looks like there
>  is some support for this but I can't find any examples.

Yes.  Take a look at scales_test.py in the examples/ directory.  This  
is using some new ticking and labelling code that I wrote for both  
"regular" axes as well as calendar axes.    The key things you need  
to do are:

from enthought.chaco2.scales.api import CalendarScaleSystem
from enthought.chaco2.scales_tick_generator import ScalesTickGenerator

# There is a new PlotAxis that uses scales, but it's not the default
# one yet.  So, you have to import it manually:
from enthought.chaco2.scales_axis import PlotAxis

# cut-n-paste the definitions of add_default_axes() from  
plot_factory.py,
# because we need it to use the PlotAxis imported above (instead of
# the one in axis.py).
#
def add_default_axes(...):
     ...

# create the axes; set the horizontal axis to use the calendar scale
left, bottom = add_default_axes(myplot)
bottom.tick_generator = ScalesTickGenerator(scale=CalendarScaleSystem())


-Peter
_______________________________________________
Chaco-users mailing list
Chaco-users@[...].com
https://mail.enthought.com/mailman/listinfo/chaco-users