[AstroPy] API question: Instantiating of time/coord and similar

[Tim Jenness]

On Wed, May 2, 2012 at 2:32 PM, Wolfgang Kerzendorf
<wkerzendorf at gmail.com> wrote:
> Hi Tim,
>
> So I'm not quite getting what you're telling us? How do you want us to implement coordinates (not WCS) into astropy (the API - not the underlying mechanics)?
>

I not trying to tell you how to do anything although I'd be happy to
help. David and I were trying to explain the concepts behind our
system. Obviously from my point of view there is 20 years of
experience in coordinate handling inside the pyast library and it
would be great to leverage that some how. I assume there is a goal of
only relying on numpy as the only non-pure python library so pyast
would not be acceptable but you could design a system that used AST
internally and switched it out at a later date when you wanted to go
pure python.

Even if you don't want to use AST I feel strongly that the idea of
separating frames from mappings and allowing frames and mappings to be
stacked together is incredibly powerful. It's a very flexible approach
that allows specific class knowledge to be added incrementally; time
support was not in the original AST release and neither was support
for spectral coordinates or IFUs.

If you want to see AST in use, the new version of DS9 relies entirely
upon AST for coordinate plotting and transformations.

>
> P.S.: I should also apologize as this was meant for the astropy-dev list and not for astropy (but I guess everyone who's on dev is on astropy as well).

I'll sign up. I had missed the dev list.

-- 
Tim Jenness
Joint Astronomy Centre.