= DataObject = '''ToDo''': the getMetadata method should probably return some kind of RDF statement container (the same interface that will be used in Extractor) instead of a Map with key-value pairs, the keys of which are specific to the type of !DataObject/!DataSource. Leo about TODO: ok with me to return an RDFMap ChangeLog: * 'Metadata' is one word, hence getMetadata, not getMetaData. * TODO: what is the semantics and use of getParent() and getChildren()? == Java Interface == {{{ #!java /** * A general interface for data objects. A data object consists of an identifier, * binary content and metadata. The object is used primarily to extract * information from datasources. For the extraction, both the InputStream * returned by getContent() and * the RDF metadata returned by getMetadata() are important. * In structured data sources that are not file-based, the getContent() method * will return null, and all structured data of the object are represented * in the getMetadata object. * */ public interface DataObject { /** * Gets the data object's primary identifier. * * @return An identifier for this data object. */ public URI getID(); /** * Gets the DataSource from which this DataObject conceptually originated. * * @return The DataSource from which this DataObject conceptually originated. */ public DataSource getDataSource(); /** * Gets the data object's parent, if any. * * @return the parent DataObject, or null when this DataObject has no parent. */ public DataObject getParent(); /** * Gets the data object's children, if any. This may be null to indicate that there * are no children. */ public Iterator getChildren(); /** * Get the source-specific metadata and data. * The used keys and values and implementation-dependent. * For java1.4 compability reasons, the map is untyped. * It is already titled RDFMap to reflect our ideas regarding RDF * * @return The scheme-specific metadata. */ public RDFMap getMetadata(); } }}}