Appendix A
The following table shows the directory hierarchy of the example.
...
- [F16H0] This is a list of the example subdirectories.
- [F16H1] This is where the accumulator files are found. Per0.xml for people and Org0.xml for organizations. In the case of this example, there are only a few entries in each of these files. In vivo.cornell.edu there are tens of thousands of people and thousands of organizations. These files are constructed by SPARQL query and XSLT. Since this is really a sideshow, it won’t be covered in detail here (see Appendix E for the SPARQL query and XSLT).
- [F16H2] The Perl script nuno is a utility for creating sequences of unique numbers and is used to generate an XML containing local names used to construct URIs for unresolved people and organization URIs. The script is self-describing when invoked as ‘nuno u’ on the bash shell command line.
- [F16H3] This directory is where RDF is placed when produced. There are files for new people and organizations and of course for thevivo:EducationalTraining instances (ED.rdf).
- [F16H4] The command sequence that produces the RDF is found here.
- [F16H5] This directory contains the result set as an XML in the file EduRS.xml.
- [F16H6] This directory contains the intermediate XML files produced along the way to the RDF.
- [F16H7] This directory contains the XSLTs used to transform EduRS.xml into ED.rdf. Notice that the file saxon9he.jar is the transform engine that executes the XSLTs.