[OSRS-PROJ] On Proj4...

[Glynn Clements]

Paul Selormey wrote:

> > To my mind, the "ultimate" solution would be to formulate the
> > projection functions as high-level symbolic expressions, to allow:
> > 
> > a) automatic conversion to any programming language, and
> > b) higher-order operations, primarily differentiation.
> 
> Any concrete suggestions in this regard? Anything you have in mind will
> help.

Well, the only real requirements would be that the format should be
easier to parse than C-with-macros, and devoid of any assumptions of
language-specific semantics. In particular, it would be nice to avoid
the use of mutable state, so that functional languages could be
supported easily.

The bulk of the work would be in analysing all of the PJ_* files to
determine the necessary feature set.

Off the top of my head, I suppose that you would need mechanims to:

a) declare projection parameters
b) declare constants
c) declare temporary variables
d) define the actual equations
e) indicate success/failure
f) supply projection metadata

-- 
Glynn Clements <glynn.clements at virgin.net>
----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.