Hi,
I was playing with schema.xml overrides and found out a couple things:
- when trying to override the geometry type of a feature type that has
a generic GeometryPropertyType I got back a Q1:PointPropertyType.
Investigation showed that it was happening because I did not
add the xmlns:gml="http://www.opengis.net/gml" in my complexType
element.
Very reasonable, yet an error, or an implicit import of the gml
namespace if missing, could be helpful
- I then tried to override the geometry type of a geometry whose
type is already know, but it's supposedly wrong. That did not
work, the original geometry type resisted my override efforts.
Not sure if this to be expected thought, as simple types
can be overridden at will (can take a string and force it into
a xs:integer... thought in output I get xs:long)
We should probably also open a jira about documenting these
schema override features.
Cheers
Andrea
--
Andrea Aime
OpenGeo - http://opengeo.org
Expert service straight from the developers.
Andrea Aime wrote:
Hi,
I was playing with schema.xml overrides and found out a couple things:
- when trying to override the geometry type of a feature type that has
a generic GeometryPropertyType I got back a Q1:PointPropertyType.
Investigation showed that it was happening because I did not
add the xmlns:gml="http://www.opengis.net/gml" in my complexType
element.
Very reasonable, yet an error, or an implicit import of the gml
namespace if missing, could be helpful
- I then tried to override the geometry type of a geometry whose
type is already know, but it's supposedly wrong. That did not
work, the original geometry type resisted my override efforts.
Not sure if this to be expected thought, as simple types
can be overridden at will (can take a string and force it into
a xs:integer... thought in output I get xs:long)
Hmm... I tried the same test and it worked for me in PostGIS. Can you include the schema.xsd you are using.
We should probably also open a jira about documenting these
schema override features.
Indeed. It seems that this feature has been brought to the forefront much more than initially expected.
Cheers
Andrea
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.
Andrea Aime wrote:
Hi,
I was playing with schema.xml overrides and found out a couple things:
- when trying to override the geometry type of a feature type that has
a generic GeometryPropertyType I got back a Q1:PointPropertyType.
Investigation showed that it was happening because I did not
add the xmlns:gml="http://www.opengis.net/gml" in my complexType
element.
Very reasonable, yet an error, or an implicit import of the gml
namespace if missing, could be helpful
Yeah, this can probably be fixed. The implicit namespace does happen, it is just there is no mapping. I think just a simple check to ensure that a namespace mapping exists should fix that issue. Filed a jira for this:
http://jira.codehaus.org/browse/GEOS-2937
- I then tried to override the geometry type of a geometry whose
type is already know, but it's supposedly wrong. That did not
work, the original geometry type resisted my override efforts.
Not sure if this to be expected thought, as simple types
can be overridden at will (can take a string and force it into
a xs:integer... thought in output I get xs:long)
We should probably also open a jira about documenting these
schema override features.
Cheers
Andrea
--
Justin Deoliveira
OpenGeo - http://opengeo.org
Enterprise support for open source geospatial.