Justin, what happens if you are trying to use an xlink:href
ClientProperty on a nillable minOccurs=0 property to encode it by reference?
For example, you might want to encode a GeoSciML gsml:GeologicUnit with
a gsml:occurrence by reference (nillable and minOccurs=0). XPath.java is
used to create this empty placeholder node. Later, a ClientProperty is
used to set xlink:href to some URI in the UserData for the node, to be
unpacked at encode time. I expect your patch would break this.
We have a unit test (GS XlinkWfsTest) that covers a gsml:MappedFeature
with a gsml:specification by reference, but that gsml:specification has
minOccurs=1 (and maxOccurs=1), so although it is nillable, it will not
be affected by your patch.
I think we need another unit test. :-)
Justin Deoliveira wrote:
> Hi all,
>
> I spent some time yesterday trying to port the geoserver wfs 1.1 xlink
> tests to an app-schema configuration. Made some progress but ran into
> the following issue.
>
> One of the types for testing looks like this:
>
> <xsd:complexType name="PrimitiveGeoFeatureType">
> <xsd:complexContent>
> <xsd:extension base="gml:AbstractFeatureType">
> <xsd:sequence>
> <xsd:element name="surfaceProperty" minOccurs="0"
> type="gml:SurfacePropertyType" />
> <xsd:element name="pointProperty" minOccurs="0"
> type="gml:PointPropertyType" />
> <xsd:element name="curveProperty" minOccurs="0"
> type="gml:CurvePropertyType" />
>
> ...
>
> Now if I understand xml schema correctly this means the element
> "surfaceProperty" is non-nillable, but there is allowed to be zero
> occurrences in them.
>
> So in the underlying database, there are records of PrimitiveGeoFeature
> which have a null value for "surfaceProperty". The current behavior that
> I am running into is that an Attribute instance is getting created for
> it, with a value set to null.
>
> However, when this gets validated an error occurs, because the type is
> non-nullable. So I would think in this case the attribute instance
> should just be omitted since minOccurs == 0, and its value is
> non-existent or null. Thoughts?
>
> If I am right does the following make sense when creating attributes
> from null values which correspond to xml elements / particles:
>
> * nillable = false, minOccurs > 0 -> create attribute with null value
> which will lead to a failure on validation
>
> * nillable = true, minOccurs > 0 -> create attribute with null value
> which is ok
>
> * nillable = false, minOccurs = 0 -> do not create an attribute
>
> Thoughts? I also whipped up the following patch, not sure if it is
> complete though:
>
> Index: app-schema/src/main/java/org/geotools/data/complex/filter/XPath.java
> ===================================================================
> --- app-schema/src/main/java/org/geotools/data/complex/filter/XPath.java
> (
> revision 33570)
> +++ app-schema/src/main/java/org/geotools/data/complex/filter/XPath.java
> (
> working copy)
> @@ -611,6 +611,14 @@
> + " is not a valid location path for type
> " + parent
> Type.getName());
> }
> int index = currStep.getIndex();
> +
> + //check for null, if the value is null but the
> descriptor is non
> -nullable
> + // then check for minOccurs = 0, and if so
> + if ( value == null && !currStepDescriptor.isNillable() ) {
> + if ( currStepDescriptor.getMinOccurs() == 0 ) {
> + return null;
> + }
> + }
> Attribute attribute = setValue(currStepDescriptor, id,
> value, in
> dex, parent,
> targetNodeType, isXlinkRef);
> return attribute;
>
> -Justin
>
--
Ben Caradoc-Davies <[email protected]>
Software Engineer, CSIRO Exploration and Mining
Australian Resources Research Centre
26 Dick Perry Ave, Kensington WA 6151, Australia
------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge
This is your chance to win up to $100,000 in prizes! For a limited time,
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
