what GeoServer version are you using? Is this an app-schema type? If so, what is your schemaUri?
Why would you want any sort of WFS import in a WFS DescribeFeatureType response? The response should be a XSD document, and unless your feature types are defined in terms of WFS (which I would find very strange), there should be no use of the WFS schema in the payload information model.
Clients should never see classpath JAR file references.
Kind regards,
Ben.
On 14/12/12 22:16, Florent Gravin wrote:
Hi,
If i make a DescribeFeatureType request, in the response, i will have
hardcoded value for the schemaLocation of wfs 2.0
--
Ben Caradoc-Davies <Ben.Caradoc-Davies@anonymised.com>
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre
Why would you want any sort of WFS import in a WFS DescribeFeatureType response? The response should be a XSD document, and unless your feature types are defined in terms of WFS (which I would find very strange), there should be no use of the WFS schema in the payload information model.
I don’t really want to have this import, but it is there i haven’t customized any thing to get this i guess.
Maybe i could try to remove it but i don’t know how to process.
Why would you want any sort of WFS import in a WFS DescribeFeatureType response? The response should be a XSD document, and unless your feature types are defined in terms of WFS (which I would find very strange), there should be no use of the WFS schema in the payload information model.
Clients should never see classpath JAR file references.
Kind regards,
Ben.
On 14/12/12 22:16, Florent Gravin wrote:
Hi,
If i make a DescribeFeatureType request, in the response, i will have
hardcoded value for the schemaLocation of wfs 2.0
–
Ben Caradoc-Davies Ben.Caradoc-Davies@anonymised.com
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre
Why would you want any sort of WFS import in a WFS DescribeFeatureType response? The response should be a XSD document, and unless your feature types are defined in terms of WFS (which I would find very strange), there should be no use of the WFS schema in the payload information model.
I don’t really want to have this import, but it is there i haven’t customized any thing to get this i guess.
Maybe i could try to remove it but i don’t know how to process.
Why would you want any sort of WFS import in a WFS DescribeFeatureType response? The response should be a XSD document, and unless your feature types are defined in terms of WFS (which I would find very strange), there should be no use of the WFS schema in the payload information model.
Clients should never see classpath JAR file references.
Kind regards,
Ben.
On 14/12/12 22:16, Florent Gravin wrote:
Hi,
If i make a DescribeFeatureType request, in the response, i will have
hardcoded value for the schemaLocation of wfs 2.0
–
Ben Caradoc-Davies Ben.Caradoc-Davies@anonymised.com
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre
We made a hack for this but i think geoserver team should have a look about this issue.
Please open a bug report and attach your patch.
Also, remember that WFS 2.0 + app schema is not an officially supported combination as far as I know,
in particular, app-schema does not work well with GML 3.2 I believe.
It may take some time to get confirmation, app-schema is developed in Australia and it’s vacation time
over there now.
Also, remember that WFS 2.0 + app schema is not an officially supported
combination as far as I know,
in particular, app-schema does not work well with GML 3.2 I believe.
All reports I have are that app-schema works with GML 3.2 (I rewrote the GeoTools GML 3.2 bindings so that it did). There are production deployments using it for GeoSciML 3 and WaterML 2.
Very little of WFS 2.0 that is not in WFS 1.1 is supported. Paging will fail silently. resolveDepth is not supported. WFS 1.1 with outputformat=gml32 works well. There are namespace prefix gotchas with known workarounds.
It may take some time to get confirmation, app-schema is developed in
Australia and it's vacation time
over there now.
--
Ben Caradoc-Davies <Ben.Caradoc-Davies@anonymised.com>
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre
Florent, I wrote that method, and I am aware that it is fragile and relies on observed behaviour of the EMF libraries, rather than reason, which does not seem to apply. A patch that fixes this problem would be welcome. A unit test would be great. Can you reproduce the problem using an existing test fixture? As I said, no response should contain a jar URL.
Kind regards,
--
Ben Caradoc-Davies <Ben.Caradoc-Davies@anonymised.com>
Software Engineer
CSIRO Earth Science and Resource Engineering
Australian Resources Research Centre