Overview

• Home
• Overview
• Download
• Publications
• Contact
• Statistics
• Acknowledgements

Converters

• Getting started
• Getty AAT, ULAN
• Geonames
• Louvre
• Joconde
• ICN
• Aziatische keramiek

Tutorial

• How to run
• Analyse XML
• First converter
• Custom rules
• Java snippets
• Using vocabularies
• Reports

Rules

• Create constant
• Rename resource property
• Rename literal property
• Replace value
• Sequence
• Lookup person
• Lookup place
• Lookup term
• Facet rename property
• Batch
• Use value of other path
• Use other subject

Data sources

• XML
• Databases SQL
• RDF SPARQL
• Custom formats

Best practices

• Version control
• Many files
• Sort in query
• Common XML
• Filter records
• Techy overview

Running

• tools.jar

Performance

• Use cases
• Path

XML files should be identified by a file name pattern. There may be many files per datasource. They will be processed one after another.

Example:

<ac:XmlDataSource>
   <ac:file>*.xml</ac:file>
   <ac:file>/additional/*.xml</ac:file>
</ac:XmlDataSource>

In the object rule a record separator should be used. In the following example XML tag mynamespace:record separates records, each identified by an id attribute with a unique code. If this code is absent (that is an error), then tag label is used to help to find the problematic record.

<ac:objectRules rdf:parseType="Collection">
   <ac:ObjectRule rdf:about="recordRule">
   <ac:recordSeparator>mynamespace:record</ac:recordSeparator>
   <ac:recordIdentifier>mynamespace:record[@id]</ac:recordIdentifier>
   <ac:recordInformalIdentifier>mynamespace:record/label</ac:recordInformalIdentifier>
   <ac:propertyRules>
    ...
   </ac:propertyRules>
</ac:ObjectRule>