annotate rdfpy/af-vocabulary.py @ 18:d5012016bf64 tip

added rdfpy and rdfonto directories

import rdflib
from rdflib import Graph, RDF, RDFS, plugin, URIRef, Literal, OWL, Namespace, XSD

plugin.register(
    'sparql', rdflib.query.Processor,
    'rdfextras.sparql.processor', 'Processor')
plugin.register(
    'sparql', rdflib.query.Result,
    'rdfextras.sparql.query', 'SPARQLQueryResult')

def getTools(graph, nsuri):
    return graph.query(
        """SELECT DISTINCT ?tool
           WHERE {
    	   	?x af:computedIn ?tool .
    		?x af:feature ?feature 
           }
    	   ORDER BY ?tool""",
        initNs=dict(
            af=Namespace(nsuri))
    )


url = 'http://sovarr.c4dm.eecs.qmul.ac.uk/af/'
afv = Namespace(url+'vocabulary/1.0#')
basedir = '/Users/alo/MusicOntology/features/'

afc = Namespace(url + 'features/')

af = Namespace(url+'ontology/1.0#')

cat = Namespace('http://sovarr.c4dm.eecs.qmul.ac.uk/features/')

vs = Namespace('http://www.w3.org/2003/06/sw-vocab-status/ns#')

graph = Graph()
graph.bind('afv', afv)
graph.bind('owl', OWL)
graph.bind('af', af)
graph.bind('vs', vs)
graph.bind('xsd', XSD)

graph.add(( vs['term_status'], RDF.type, OWL.AnnotationProperty ))
graph.add(( afv['computedIn'], RDF.type, OWL.Property ))
graph.add(( afv['computedIn'], RDFS.range, XSD.String ))
graph.add(( afv['computedIn'], RDFS.domain, af['AudioFeature'] ))

source = Graph()
source.parse(basedir + 'af-catalogue-01.rdf')

tooldict = {}

for tool in getTools(source, cat):
    idt = afv[tool[0]]
    tooldict[tool[0]] = idt
    graph.add((idt, RDF.type, OWL.Class))
    graph.add((idt, RDFS.label, tool[0]))
    graph.add((idt, RDFS.comment, tool[0] ))
    graph.add((idt, vs['term_status'], Literal("testing", lang="en") ))

for sub in source.subjects(RDF.type, OWL.Class):
    name = sub.split('/')[-1]
    id = name.replace("(", "").replace(")", "").replace("4Hz", "")
    #id = ''.join(i for i in id if not i.isdigit())
    id = URIRef(afv+id)
    graph.add((id, RDF.type, OWL.Class))
    graph.add((id, RDFS.subClassOf, af['AudioFeature']))
    graph.add((id, vs['term_status'], Literal("testing", lang="en") ))
    count = sum(1 for _ in source.objects(sub, afc['feature']))
    if count > 0:
        labelAdded = False
        for label in source.objects(sub, afc['feature']):
            if not labelAdded:
                graph.add(( id, RDFS.label, Literal(label, lang="en") ))
                graph.add(( id, RDFS.comment, Literal(label) ))
                labelAdded = True
    else:        
        if name.find("ADRess") > -1:
            graph.add(( id, RDFS.label, Literal(name + " (Azimuth Discrimination and Resynthesis)", lang="en") ))
            graph.add(( id, RDFS.comment, Literal(name + " (Azimuth Discrimination and Resynthesis)") ))
        else:
            if name.find("Aim") == 0:
                graph.add(( id, RDFS.label, Literal(name + " (Auditory Image Model) ", lang="en") ))
                graph.add(( id, RDFS.comment, Literal(name + " (Auditory Image Model) ") ))
            else:
                graph.add(( id, RDFS.label, Literal(name, lang="en")))
                graph.add(( id, RDFS.comment, Literal(name) ))
    
    for tool in source.objects(sub, cat['computedIn']):
        graph.add((id, afv['computedIn'], tooldict[tool]))
    
graph.serialize(basedir+'rdfonto/af-vocabulary.rdf')
graph.serialize(basedir+'rdfonto/af-vocabulary.n3', format="n3")