[Kst] any advice on datasource access to current plot range?

[Ted Kisner]

On Sunday 06 November 2005 20:29, George Staikos wrote:
>   This is definitely possible.  Curves are just siblings of all other data
> objects that happen to use data objects.  It's going to require ongoing
> genericizing of kst2dplot and other objects though.
>
> > In the mean time, any suggestions for temporarily hacking in this kind of
> > support are welcome...
>
>   I'm not sure at the moment.

ok, this is pretty much what I thought- just wanted to hear from the experts 
to see if I was missing something.  

I think that for 1.2, healpix is currently as good as it gets.  I might add 
some more to the testcase, but probably not for a while.

For 1.3+, the things I would like to work on are:

A)  abstraction of the rendering to a separate class (Kst2DRenderer), so that 
all the basecurves and Kst2DPlot can just worry about doing things in generic 
(U,V) coordinates without dealing with the details of the current projection.

B)  create a new generic KstVectorField basecurve that takes 4 KstVectors and 
plots a vectorfield.

C)  either modify KstImage to also take a vector of healpix values and draw 
it, or create a new KstHealpix basecurve that is very similar to KstImage, 
but which takes a vector of pixel values and draws a colormap.

As part of B and C, we might want to make KstBaseCurve a plugin architecture 
and rework the datamanager dialog to have 2 comboboxes (rather than a set of 
pushbuttons).  One combobox should let the user "create a new data 
object" (vector, matrix, equation, plugin, event monitor) and the other 
should let the user "create a new data display object" (curve, image, 
histogram, power spectrum, csd).

Anyway, I'll start working on expanding these proposals by putting stuff 
in /devel-docs and /patches.

cheers,

-Ted