[mapguide-users] MapGuide 2.X : Why Developers don't develop MapGuide to replace the local arbitrary SRS of DataSource by the SRS of the parent Map when we open a session ?

[Jackie Ng]

I hope you understand what an *arbitrary* SRS means. 

It's an SRS that has no real world context. The coordinates of your
geometries cannot be determined on the surface of planet earth.

Now for a Map to do its auto-magical re-projection of layers, the Map's
coordinate system and the layer's coordinate system must be of the same
type. It can't re-project a layer if it's data is based on arbitrary
coordinates. 

Think of what I already said: Geometries in arbitrary SRS coordinates cannot
be determined on the Earth's surface, so how could CS-Map possibly know what
real world context from which to transform said layer's coordinates to the
coordinate system of your Map?

Okay, suppose your Feature Source has an arbitrary SRS, but you know what
SRS the coordinates actually represent. Then this is where you can use the
Coordinate System override mechanism in the Feature Source to replace the
arbitrary SRS with a non-arbitrary one, allowing for the Map to re-project
such data if required. The Coordinate System override is a hint to MapGuide
as to how to interpret the coordinates from that Feature Source. It won't do
any validation. For example, if you override your SRS with WGS84, MapGuide
is not going to check that your coordinates must be within [-180, 180],
[-90, 90].

If you're intentionally using your *geospatial* data with a local SRS, I'd
really question why you must do this. Do you give SHP files to people
without the .prj file? That's effectively what you're doing.

- Jackie



--
View this message in context: http://osgeo-org.1560.n6.nabble.com/MapGuide-2-X-Why-Developers-don-t-develop-MapGuide-to-replace-the-local-arbitrary-SRS-of-DataSource--tp5031537p5031593.html
Sent from the MapGuide Users mailing list archive at Nabble.com.