Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [udig-devel] Rendering pipeline: how CRS is used.
You are on the right track. Mind you it depends on the renderer you are using. Shapefile has a optimized renderer that has a different pipeline from the Streaming renderer which is used to render other datastores. I think I explain the StreamingRenderer as it is used by more datastores.
I'm going to take it from the client code as well. I'm assuming that the geometry is either a line or a polygon. Points and GridCoverages are handled in a similar manner but the styling and drawing section is slightly different. 

1. Create a MapContext.
2. Add MapLayer(s) to context.
3. Set Area of Interest and CRS on the MapContext
4. Create renderer.
5. Set MapContext of renderer
foreach( layer ) do {
6. FeatureType is obtained from Datastore. (Contains the CRS in the *current* Feature model) 
  7.  Area of Interest is reprojected into the CRS of the layer.
  8a.  A query is constructed from the projected Area of Interest.
8b. The query is restricted to only request Attributes that are required for styling the feature. (required for thematic styling, and other non-basic styles) 9. A MathTransfrom is constructed that transforms from Layer CRS to Screen space 
  foreach feature do {
    9.  Feature is obtained from the datastore.
    9a.  (Assuming shapefile)  Geometry is created, no reprojection
    9b.  Attributes are read
    9c.  Feature is created.
10. Geometry is decimated. (Vertices that map to a single pixel are collapsed to a single vertex) 
    11. Geometry is reprojected using the constructed MathTransform
    12.  A java.awt.Shape is constructed that wraps the Geometry
13. The graphics object is set according to the style. (Fill, color, pattern, etc... are all set) 
    14. Shape is drawn.
15. If labeling is done the label for the current feature is put in a cache. 
  }
}
16.  Labels are drawn.

Hope this helps.

Jesse

On 23-May-06, at 6:23 AM, Adrian Custer wrote:


Hey all,

Could someone explain the uses of CRS and transforms in the rendering
pipeline? I need a coarse overview of rendering as used by
uDig/streaming Renderer to understand:
 (1) how we go from shapefile data to screen data
 (2) what CRS's are stored in the rendering system
This is all prep work so I can compose an intro to CRS that will work as 
background for a future, more complex discussion of CRS with features
and renderers.

What I really want is a pipeline, from a UTM shapefile (as an example)
to a displayed map on screen where we know what CRS's are in play and
where the transformations are performed. The initial start of the
operations are:

        10) Get Shapefile Metadata   => CRS used for
        geometryAttributeType
        20) Get Shapefile coordinates
        30) Build a geometry
        40) Build a feature
        ...
        ?) Create a Renderer (? with a crs?)
        ?) add the feature to the chain
        ...
        ?) transform to projected coordinates (if in lat/long?) ?
        ...
        ?) transform to image coordinates
        ...
        ?) transform to screen coordinates
If someone could flesh out the chain so I have a working idea of what's 
going on, I would appreciate it.

thanks,
adrian

_______________________________________________
User-friendly Desktop Internet GIS (uDig)
http://udig.refractions.net
http://lists.refractions.net/mailman/listinfo/udig-devel

Back to the top