nothing@18: import rdflib, os, fnmatch, urllib2
nothing@18: from rdflib import Graph, RDF, RDFS, plugin, URIRef, Literal, OWL, XSD, Namespace
nothing@18: 
nothing@18: graph = Graph()
nothing@18: 
nothing@18: af = Namespace('http://sovarr.c4dm.eecs.qmul.ac.uk/features/')
nothing@18: graph.bind('local', af)
nothing@18: 
nothing@18: dc = Namespace('http://purl.org/dc/elements/1.1/')
nothing@18: graph.bind('dc', dc)
nothing@18: 
nothing@18: categories = [
nothing@18:     "Temporal",
nothing@18:     "Energy",
nothing@18:     "Spectral",
nothing@18:     "Harmonic",
nothing@18:     "Perceptual",
nothing@18:     "Various"
nothing@18: ]
nothing@18: 
nothing@18: subcategories = {
nothing@18:     "Temporal": ["Global Temporal Feature", "Instantaneous Temporal Feature"],
nothing@18:     "Spectral": ["Spectral Shape", "Global Spectral Shape Description"],
nothing@18:     "Harmonic": ["Harmonic Spectral Shape"],
nothing@18:     "Perceptual": ["Perceptual Spectral Envelope Shape"],
nothing@18:     
nothing@18: }
nothing@18: 
nothing@18: flat = []
nothing@18: 
nothing@18: lines = [line.strip() for line in open('pdfextract/cuidado.txt')]
nothing@18: 
nothing@18: category = "Temporal"
nothing@18: subcategory = "Global Temporal Feature"
nothing@18: 
nothing@18: lineIndex = 5
nothing@18: 
nothing@18: for line in lines[5:]:
nothing@18:     addFeature = True
nothing@18:     if line.find("Features") > -1:
nothing@18:         addFeature = False
nothing@18:         if categories.count(line[0:line.find("Features")-1]) > 0:
nothing@18:             if line[0:line.find("Features")-1] != category:
nothing@18:                 category = line[0:line.find("Features")-1]
nothing@18:                 subcategory = ""
nothing@18:     for sub in flat:
nothing@18:         if line == sub:
nothing@18:             addFeature = False
nothing@18:             if subcategory != line:
nothing@18:                 subcategory = line
nothing@18:             
nothing@18:     if line.count(":") == 0 and line.count("_") == 0 and len(line) > 2 and addFeature:
nothing@18:         id = af + line.title().replace(" ", "")
nothing@18:         graph.add((
nothing@18:             URIRef(id),
nothing@18:             RDF.type,
nothing@18:             RDFS.Resource
nothing@18:         ))
nothing@18:         graph.add((
nothing@18:             URIRef(id),
nothing@18:             af['feature'],
nothing@18:             Literal(line)
nothing@18:         ))
nothing@18:         graph.add((
nothing@18:             URIRef(id),
nothing@18:             URIRef(af+"Dimensions"),
nothing@18:             Literal(lines[lineIndex+2])
nothing@18:         ))
nothing@18:                     
nothing@18:         graph.add((
nothing@18:             URIRef(id),
nothing@18:             URIRef(af+"FrameBased"),
nothing@18:             Literal(lines[lineIndex+1])
nothing@18:         ))
nothing@18:         
nothing@18:         if subcategory == "":
nothing@18:             graph.add((
nothing@18:                 URIRef(id),
nothing@18:                 af['typeOfFeature'],
nothing@18:                 Literal(category + " Feature")
nothing@18:             ))
nothing@18:         else:
nothing@18:             graph.add((
nothing@18:                 URIRef(id),
nothing@18:                 af['typeOfFeature'],
nothing@18:                 Literal(subcategory.replace(" ", ""))
nothing@18:             ))
nothing@18: 
nothing@18:     lineIndex += 1
nothing@18:             
nothing@18: 
nothing@18: graph.serialize('/Users/alo/MusicOntology/features/rdf/af-cuidado.rdf')
nothing@18: graph.serialize('/Users/alo/MusicOntology/features/rdfn3/af-cuidado.n3', format="n3")