[geotk] Re: Gaus Conform Projection

[Martin Desruisseaux]

theuns a écrit :
> Any chance of the projection being implemented in Geotk?

It depends partially on volunter time. The most tedious part is to write a test 
suite. I can help, but chances to see that projection in raisonable time frame 
are greater if a volunter can do the following steps:

1) Find the formula. The following web site provides C methods which can
    be used as a starting point (if Frank Wamardam give us permission) if
    that projection can be found in the list (sometime with a different
    name):

    http://www.remotesensing.org/geotiff/proj_list/

2) Copy-and-paste an existing class from Geotk (I suggest
    org.geotoolkit.referencing.operation.projection.Mercator)
    and replace the old formula by the new one.

3) Copy-and-paste an existing test class from Geotk
    (e.g. MercatorTest) and replace the tested coordinates by
    the one appropriate for the new projection. At least some
    of the expected projected coordinates should be computed
    by an other software (e.g. ESRI or Proj4).

 From that point I can review those two classes, write the corresponding 
provider class and commit them. Then there is a final step:

4) Create a new test script similar to Mercator.txt for the new
    projection. It is not necessary to have as many test points than
    Mercator, but 10 points would be nice.

	Martin