We are using ST_Geometry for Oracle 11g at the moment.
And we need to query features (points,lines,polygons in a view) inside a given extent(either a circle or a polygon or a rectangle), and we use the st_envintersects to do the spatial calculation.
However since some of the features are poygons which make the query rather slowly even we create indexes for the geometry column of each feature.
For example there is a view feature LView with columns:
objectid lon( lat geom(the geometry, point or polygon or polyline) ... And we use the following sql to do the query:
select * from Landuse l where sde.st_envintersects(l.geom,sde.st_geometry('POLYGON((....))',3)) = 1; It will take almost 2-3 minutes to do the querying and there is only 2000 records in the view.
Then when we re-create the LView and remove the features of polyon and polylines whose geometry is too complex.
The same query will take only a few seconds.
It seems that the spatial calculation for complex the polygon will slow down the query. So I want to treat the polygons as points(by the lon lat column) during the query.
Something like this:
select * from Landuse l where sde.st_envintersects(sde.st_geometry('Point(l.lon,l.lat)',3),sde.st_geometry('POLYGON((....))',3)) = 1; Is this possible
أكثر...
And we need to query features (points,lines,polygons in a view) inside a given extent(either a circle or a polygon or a rectangle), and we use the st_envintersects to do the spatial calculation.
However since some of the features are poygons which make the query rather slowly even we create indexes for the geometry column of each feature.
For example there is a view feature LView with columns:
objectid lon( lat geom(the geometry, point or polygon or polyline) ... And we use the following sql to do the query:
select * from Landuse l where sde.st_envintersects(l.geom,sde.st_geometry('POLYGON((....))',3)) = 1; It will take almost 2-3 minutes to do the querying and there is only 2000 records in the view.
Then when we re-create the LView and remove the features of polyon and polylines whose geometry is too complex.
The same query will take only a few seconds.
It seems that the spatial calculation for complex the polygon will slow down the query. So I want to treat the polygons as points(by the lon lat column) during the query.
Something like this:
select * from Landuse l where sde.st_envintersects(sde.st_geometry('Point(l.lon,l.lat)',3),sde.st_geometry('POLYGON((....))',3)) = 1; Is this possible
أكثر...