Previous topic  Top  Next topic  Print this Topic


frageWhat is the difference between RDFTerm-equal and sameTerm?

Comparing two terms by equality (operator '=') checks for value equality, e.g. "1.0"^^xsd:double is equal to "1"^^xsd:int. sameTerm in contrast checks for identity, e.g. "1.0"^^xsd:double is different from "1"^^xsd:int. Please note that due to the implicit type casting e.g. "1"^^xsd:long and "1"^^xsd:int will be identical (and of course equal too). Depending on your data, if the check sameTerm is sufficient, a huge performance benefit might be the result. Especially in combination with a persistent data model sameTerm is much more efficient.


frageDefault and named graphs?

The dataset is defined by the query using a combination of FROM and FROM NAMED. If one or more named graphs are specified but no default graph the default graph is empty. If only one or more default graphs are given but no named graphs the set of named graphs is empty. Finally, if neither a default nor named graphs are specified the default dataset is used. This contains the current, i.e. queried, ontology as a default graph and all other graphs as named graphs.


frageObjectLogic support?

SPARQL functionality might also be used in combination with a ObjectLogic data model. Care has to be taken for the slightly different axiomatization of ObjectLogic and differences in handling/representation of property domain/range information. Therefore the result of SPARQL queries might differ. For loading new ontologies via SPARQL Update the ontology format has to match the data model ontology language, e.g. for ObjectLogic only ObjectLogic ontologies can be loaded. Finally if non-RDF-terms like functional terms are returned by SPARQL queries these are converted to special URIs (obl:term#ObjectLogic) similar to the ontology transformation.


frageNames ground optimization?

Depending on the chosen names ground configuration not all of the SPARQL/RDF features are available. A safe option without any restriction is to activate module names ground. This might increase the overall performance. If you choose concept names ground or attribute names ground the merging of the default graph in SPARQL queries is not supported. In addition, no sub-properties may be used for attribute names ground. Those single restrictions also apply for the combination of different names ground settings.