[Proj] Some questions

[support.mn at elisanet.fi]

Frank Warmerdam [warmerdam at pobox.com] kirjoitti: 

> There is no immediate plan to re-engineer PROJ.4 to use libproj4 as it's
> underlying projections library though that has been discussed as a
> possibility.
> 

Ok.

> The double ellipsoid case was not really resolved.  It is still
> necessary in some cases to use tricks like +nadgrids=@null to
> effectively treat a non-WGS84 ellipsoid based dataset as if it were
> based on WGS84 as a datum.
> 
> The differences in ellipsoid can be very significant.  The most
> common situation this comes up is with sphere based projections
> (like google spherical mercator) where the projection is on a
> sphere but you are expected afterwards to treat the lat/long
> result as being on WGS84.  In this situation incorrect interpretation
> can result in very substantial errors.
> 

Is there any easy way to open the projection and datum shift ellipsoids
for the end user to be altered in those cases where required?

For example something like:

pellps=.....
dellps=.....

where "pellps" would stand for "projection ellipsoid" and
"dellps" for "datum ellipsoid".

and when both are the same, just normal

ellps=....

where "ellps" stands for both datum and projection. That
would not alter the existing definitions, but allow more
accurate control of the ellipsoids involved.

Also the other ellipsoid parameters would get same type
of modifiers... p and d.

The idea would be to allow the proj.4 definition to have full
control of the calculations involved. Maybe some day in the
future?

Janne.