Information Technology Reference
In-Depth Information
-
(APIClassNodeA ihneritsEdge APIClassNodeB)
→
(OwlNodeB subclassOf
OwlNodeA)
If a class A inherits a class B in the API graph, the relation becomes a
subclassOf relation between the correspondent classes of the OWL graph.
-
(APIClassNodeA hasAttributeEdge APIClassNodeB )
→
(OwlNodeA Ob-
jectProperty: hasProperty OwlNodeB)
If a class A has an attribute B in the API graph, the relation becomes
an Object Property with label ”hasProperyy” between the correspondent
classes of the OWL graph.
-
(APIClassNodeA hasMethodEdge APIClassMethodNodeB)
→
(OwlNodeA
ObjectProperty: isDoer OwlNodeB)
If a class A has a method B in the API graph, the relation becomes an
Object Property with label ”isDoer” between the correspondent elements of
the OWL graph.
-
(APIClassMethodNodeA hasMethod APIClassMethodNodeConstructorB
andAPIClassMethodNodeConstructorBhasInputParameterAPIParameterN-
odeC)
(OwlNodeClassA ObjectProperty: hasProperty OwlClassC)
If a class A has a constructor with some input parameters, in the owl graph
there are Object Properties with label ”hasProperty” between the corre-
spondent elements.
-
(APIClassMethodNodeA hasInputParameterEdge APIParameterNodeB)
→
→
(OwlNodeA ObjectProperty: actsOn OwlNodeB )
If a method A has some input parameter, there are Object Properties with
label ”actsOn” between the correspondent elements on the OWL graph.
-
(APIClassMethodNodeA hasReturnTypeEdge APIClassNodeB)
→
(OwlN-
odeA ObjectProperty: produce OwlNodeB )
If a method A has a return type B there is an Object Property with label
”produce” between the correspondent elements in the OWL graph.
5Con lu on
In this paper we have proposed an approach to perform automatically, or with
automated support, operations like the alignment and mapping of software which
will be useful to perform software modernization and migration. The method-
ology is based on an automatic analysis and representation of code at higher
level of abstraction than the syntactical and structural one that enables auto-
matic recognition of the algorithms and algorithmic concepts implemented in
thesourcecode.Basedonmatchmakingtechniques, the concepts recognized are
compared with functional concepts represented by the ontologies and the results
provide useful information to perform porting of source code excerpts and API
calls to the target cloud programming environment. The architecture support-
ing this methodology is composed of four components: the Algorithmic COncept
Recognizer, which recognizes algorithmic concepts in the code, the API Ontology
Builder, which extracts the graph by parsing source code and produces an ontol-
ogy graph applying graph transformation patterns, the Schema Matcher which
Search WWH ::
Custom Search