WCS Technical Description
How the WCS works
Just enough information to be dangerous ...
WCS
The WCS 1.1.2 software implements a WCS service that is close to generic. It can work using a simple XML catalog that lists WCS coverages ('coverages' hereafter), at lest in theory. This software using the semantic.wcs package to perform actions like access coverages and rebuild the catalog of served coverages.
semantics.wcs
The software in semantic.wcs is the interface to the 'semantic engine' that the IRISail software implements. It doesn't know anything about inferencing per se, but it implements an interface to the high-level operations provided by the inferencing software. All of the code is contained in one class called StaticRdfCatalog that has a number of public methods that return coverages and other useful information about coverages. It also contains an update() method used to (re)build the contents of the semantic repository and then the coverages that can be derived from the information in that repository. The end result of running the update() method is that an XML file listing the known coverage descriptions is written into the content/WCS/StaticRdfCatalog/coverageXMLfromRDF.xml file. This file is then used to build an in-memory representation of WCS coverages that the other methods of the StaticRdfCatalog class use.
IRISail
This is the semantic engine. It uses information about WCS and other metadata standards to build coverages for WCS using whatever metadata is available. It leverages ontologies that describe the different metadata standards & conventions in general terms.
The entry point for the inferencing software is the class RespoitoryOps.
The inferencing software uses the sesame software for both inferencing and as a tripple store for information, both initial conditions and derived information.
