Â¥Ö÷

[ÇóÖú]com.hp.hpl.jena.graph.query¿ÉÒÔʵÏÖÄÇЩ²éѯ¹¦ÄÜ£¿ ÇëÎÊcom.hp.hpl.jena.graph.query¿ÉÒÔʵÏÖÄÇЩ²éѯ¹¦ÄÜ£¿    Êղؠ  ·ÖÏí   ¶¥(0)

	µÚ2Â¥

ʲô½ÐÄÄЩ¹¦ÄÜ£¿

	µÚ3Â¥

ÒÔÏÂÊÇÒýÓÃjiexincaoÔÚ2007-4-20 8:51:00µÄ·¢ÑÔ£º ʲô½ÐÄÄЩ¹¦ÄÜ£¿ ¼ÈÈ»ÓÐÁËARQ£¬ÎªÊ²Ã´»¹ÐèÒªÕâ¸ö°üѽ£¿

	µÚ4Â¥

Ö»ÌṩÁËÒ»¾ä»°µÄÃèÊö°¡ Package com.hp.hpl.jena.graph.query Description Defines the graph-level Query classes, allowing multi-pattern query over arbitrary graphs.

	µÚ5Â¥

com.hp.hpl.jena.graph.query ÊÇgraph-level²ãµÄ ¶øARQÊÇmodel-level²ãµÄ ¹ØÓÚgraph-levelºÍmodel-levelµÄÇø±ðÓëÁªÏµ£¬Çë¿´£º 2. JENA2 ARCHITECTURE OVERVIEW The heart of the Jena2 architecture is the RDF graph, a set of triples of nodes. This is shown in the Graph layer (see figure 1). This layer, following the RDF abstract syntax, is minimal by design: wherever possible functionality is done in other layers. This permits a range of implementations of this layer such as inmemory or persistence triple stores. The EnhGraph layer is the extension point on which to build APIs: within Jena2 the functionality offered by the EnhGraph layer is used to implement the Jena1 Model API and the new Ontology functionality for OWL and RDFS, upgrading the Jena1 DAML API. I/O is done in the Model layer, essentially for historical reasons. The Jena2 architecture supports fast path query that goes all the way through the layers from RDQL at the top right through to an SQL database at the bottom, allowing user queries to be optimized by the SQL query optimizer. We give some more detail on the three layers below. 2.1 The Graph Layer: Triples as the Universal Data Structure The Graph layer is based on the RDF Abstract Syntax [20]. It is straightforward to implement: ¡¤ triple stores, both in memory and backed by persistent storage; ¡¤ read-only views of non-triple data as triples, such as data read from a computer file system hierarchy, or scraped from a web page; ¡¤ virtual triples corresponding to the results of inference processes over some further set of triples as premises. Implementations of the Graph layer provided with Jena2 give a variety of concrete (materialized) triple stores, and some built-in inference: specifically for RDFS and for a subset of OWL. 2.2 The Model Layer: Views for Application Programmers Jena2 maintains the Model API from Jena1 as the primary abstraction of the RDF graph used by the application programmer. This gives a much richer set of methods for operating on both the graph itself (the Model interface) and the nodes within the graph (the Resource interface and its subclasses). Further, the DAML API is updated and enhanced in Jena2 to form an Ontology API that can be realized as a DAML API or an OWL API.

	µÚ6Â¥

thanks!

	µÚ7Â¥

¸Ð¾õÀàËÆÓÚXML SchemaºÍXML DocµÄ¹Øϵ£¬ÊÇÕâÑùµÄÂð£¿