I cann't answer how close we are if we patch the schemas to handle ANY_ELEMENT... I think it will change the way ELEMENT is being handled in application schema since ELEMENT_ELEMENT is an XSDElementDeclaration object and ANY_ELEMENT is an XSDWildcard object.
-----Original Message----- From: Justin Deoliveira [mailto:jdeol...@opengeo.org] Sent: Friday, 14 May 2010 11:31 PM To: Tan, Florence (CESRE, Kensington) Cc: Caradoc-Davies, Ben (CESRE, Kensington); geotools-devel@lists.sourceforge.net Subject: Re: [Geotools-devel] Encode using XSD with <any/> element While I can't confirm Florence's conclusion sounds very reasonable. It is most likely GeoTools not handling xs any content. So if we patch schemas to handle ANY_ELEMENT does that bring us any closer? On 10-05-13 11:19 PM, florence....@csiro.au wrote: > Hi Ben& Justin, > > I think I might be wrong of saying the EMF model not handling<any/> element, > as when I step thru the code at org.geotools.xml.Schemas (breakpoint at line > 831), I can see the particle has an xs:any element > (org.eclipse.xsd.impl.xsdparticlei...@f16568 (element: [xs:any: null]) > (minOccurs: 0, maxOccurs: -1)), but Schemas class will skip it since it > doesn't handle "ANY_ELEMENT". So is it correct to say that it is not handled > in geotools schemas parsing ? > > Regards, > Florence > > -----Original Message----- > From: Caradoc-Davies, Ben (CESRE, Kensington) > Sent: Friday, 14 May 2010 11:13 AM > To: Justin Deoliveira > Cc: Tan, Florence (CESRE, Kensington); geotools-devel@lists.sourceforge.net > Subject: Re: [Geotools-devel] Encode using XSD with<any/> element > > Justin, > > do you know if the version of EMF we are using (ecore-2.2.2 and > xsd-2.2.2) even support the<any/> element as a content model? If so, > how should it appear in the XSDTypeDefinition? Florence reports seeing > no elements in the EMF model of the parsed schema. > > Kind regards, > Ben. > > > On 12/05/10 22:31, Justin Deoliveira wrote: >> It is tough to say but doubtful. Either XSAnyTypeBinding would have to >> able to return properties from generic objects (which it does not), or >> an application schema woudl have to extend AnyOrReferenceType to >> restrict the content to some other type that the encoder does know how >> to encode. So yeah, almost certainly I would think this would not be >> handled out of the box. >> >> How do you represent such an object in the geotools feature model? Once >> you have some way of representing these things I think you will have to >> implement a special binding for it that can handle all cases. Sort of >> the same way the AbstractFeatureType bindings work. >> >> Hope that helps. >> >> -Justin >> >> On 10-05-11 11:15 PM, florence....@csiro.au wrote: >>> Hi Justin, >>> Do you expect Encoder to be able to encode complexType (defined in an >>> application schema) like this : >>> <complexType name="AnyOrReferenceType"> >>> <sequence minOccurs="0"> >>> <any/> >>> </sequence> >>> <attributeGroup ref="gml:AssociationAttributeGroup"/> >>> </complexType> >>> Regards, >>> Florence > > -- Justin Deoliveira OpenGeo - http://opengeo.org Enterprise support for open source geospatial. ------------------------------------------------------------------------------ _______________________________________________ Geotools-devel mailing list Geotools-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
