Versions Compared


  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 5.3


Remaining Issues: (see Feature Model Design Discussion for explaination of options considered)

  • Facets, as Filter or Expression or something else?
  • Associations, do we need an explict representation?


Q: We have a question - is Geometry Complex?
A: No.


It would be desirable to:

  • allow the testing of individual values prior to their being assembled into a completed Feature
  • generalize Filter to accept other kinds of XPathable content

Other toolkits have taken to including the bare minimum needed to get shapefile working (maxLength and so on). It would be great if we can figure out a way to make use an Expression syntax to solve a general need:


You can see that the intended separation is maintained by both GeoTools and this Proposal.
It should be noted that XPath works off the Type system, validation works of the Reference system.

At first draft we had:






Object, Complex, Feature, FeatureCollection

Attribtue, Feature

Object, Feature, FeatureCollection


Schema, ...

Schema, ...

AttributeType, FeatureAttributeType


Type, ComplexType, FeatureType, FeatureCollectionType

AttributeType, FeatureType


The two drafts mostly differ in:

  • the direction of associations (strangely enough aligned with the authors primary concerns of Validation/Schema and XPath/Containment)
  • treatment of Atomic and Collection Types

These how now been merged.


Note on multiplicity & getAttribute w/ name or attributeType:

  • a single instance when attribute type minOccus=1 and maxOccurs=1
  • nill or single instance when attributeType minOccurs=0 and maxOccurs=1
  • List when minOccurs=0 and maxOccurs=unbounded