nothing@1: import rdflib, os, fnmatch, urllib2
nothing@1: from rdflib import Graph, RDF, RDFS, plugin, URIRef, Literal, OWL
nothing@1: from xml.dom.minidom import parseString
nothing@1: 
nothing@1: names = [line.strip() for line in open('pdfextract/names.txt')]
nothing@1: cat = [line.strip() for line in open('pdfextract/categories.txt')]
nothing@1: sig = [line.strip() for line in open('pdfextract/sig.txt')]
nothing@1: 
nothing@4: graph = Graph()
nothing@1: 
nothing@4: af = Namespace('http://sovarr.c4dm.eecs.qmul.ac.uk/features/')
nothing@4: graph.bind('af', af)
nothing@4: 
nothing@4: dc = Namespace('http://purl.org/dc/elements/1.1/')
nothing@4: graph.bind('dc', dc)
nothing@4: 
nothing@4: owl = Namespace('http://www.w3.org/2002/07/owl#')
nothing@4: graph.bind('owl', owl)
nothing@1: 
nothing@1: i = 0
nothing@1: 
nothing@1: order = [
nothing@1:     "Zero Crossing Rate", 
nothing@1:     "Linear Predictive Coding", 
nothing@1:     "Mel-scale Frequency Cepstral Coefficients", 
nothing@1:     "Auditory Filter Bank Temporal Envelopes",
nothing@1:     "Rate-scale-frequency Features",
nothing@1:     "Phase Space Features"
nothing@1: ]
nothing@1: 
nothing@1: domains = {
nothing@1:     "Zero Crossing Rate": 'temporal',
nothing@1:     "Linear Predictive Coding": 'frequency',
nothing@1:     "Mel-scale Frequency Cepstral Coefficients": 'cepstral',
nothing@1:     "Auditory Filter Bank Temporal Envelopes": 'modulation frequency',
nothing@1:     "Rate-scale-frequency Features": 'eigendomain',
nothing@1:     "Phase Space Features": 'phase space'
nothing@1: }
nothing@1: 
nothing@1: abbr = {
nothing@1: 	"Zero Crossing Rate": "ZCR",
nothing@1: 	"Mel-scale Frequency Cepstral Coefficients": "MFCC",
nothing@1:     "Linear Predictive Coding": "LPC",
nothing@1:     "Linear Prediction Cepstral Coefficients": "LPCC",
nothing@1:     "Zero crossing peak amplitudes": "ZCPA",
nothing@1:     "Line spectral frequencies": "LSF",
nothing@1:     "Short-time energy": "STE",
nothing@1:     "Amplitude descriptor":  "AD",
nothing@1:     "Adaptive time frequency transform": "ATFT",
nothing@1:     "Daubechies Wavelet coefficient histogram": "DWCH",
nothing@1:     "Spectral Flux": "SF",
nothing@1:     "Group delay function": "GDF",
nothing@1:     "Modified group delay function": "MGDF",
nothing@1:     "Spectral centroid": "SC",
nothing@1:     "Subband spectral flux": "SSF",
nothing@1:     "Perceptual linear prediction": "PLP"
nothing@1: }
nothing@1: 
nothing@1: 
nothing@1: domain = ""
nothing@1: domainIndex = 0
nothing@1: compdict = {}
nothing@1: 
nothing@1: for filename in ['filters', 'trans', 'aggr']:
nothing@1:     for line in [line.strip() for line in open('pdfextract/' + filename + '.txt')]:
nothing@1:         compdict[line[0]] = line[2:]
nothing@1:         
nothing@1: 
nothing@1:             
nothing@1: for name in names:
nothing@1:     id = local + (name.replace(' ','').replace('-',''))
nothing@1:     
nothing@1:     if name == order[domainIndex]:
nothing@1:         domain = domains[order[domainIndex]]
nothing@1:         domainIndex += 1
nothing@1:     
nothing@1:     graph.add(( URIRef(id),        
nothing@1:         URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#type'), 
nothing@1:         URIRef(OWL.Class)  
nothing@1:     ))
nothing@1:     
nothing@1:     graph.add(( 
nothing@1:         URIRef(id), 
nothing@1:         URIRef(local + 'feature'), 
nothing@1:         Literal(name.replace(' ','').replace('-','')) 
nothing@1:     ))
nothing@1: 
nothing@1:     graph.add(( 
nothing@1:         URIRef(id), 
nothing@1:         URIRef(local + 'domain'), 
nothing@1:         Literal(domain) 
nothing@1:     ))
nothing@1: 
nothing@1:     word = cat[i].split(' ')
nothing@1: 
nothing@1:     temp = {
nothing@1:         'I': 'intraframe',
nothing@1:         'X': 'interframe',
nothing@1:         'G': 'global'
nothing@1:     }[word[0]]
nothing@1: 
nothing@1:     graph.add(( 
nothing@1:         URIRef(id), 
nothing@1:         URIRef(local + 'temporalscale'), 
nothing@1:         Literal(temp) 
nothing@1:     ))
nothing@1:     
nothing@1:     if word[1] == 'Y': 
nothing@1:         temp = 'perceptual'
nothing@1:     else:
nothing@1:         temp = 'physical'
nothing@1:         
nothing@1:     graph.add((
nothing@1:         URIRef(id), 
nothing@1:         URIRef(local + 'level'), 
nothing@1:         Literal(temp) 
nothing@1:     ))
nothing@1:         
nothing@1:     if word[2] == 'Y':
nothing@1:         graph.add((
nothing@1:             URIRef(id), 
nothing@1:             URIRef(local + 'model'), 
nothing@1:             Literal('psychoacoustic') 
nothing@1:         ))
nothing@1:         
nothing@1:     temp = {
nothing@1:         'L': 'low',
nothing@1:         'M': 'medium',
nothing@1:         'H': 'high'
nothing@1:     }[word[3]]
nothing@1:     
nothing@1:     graph.add(( 
nothing@1:         URIRef(id),
nothing@1:         URIRef(local + 'complexity'), 
nothing@1:         Literal(temp) 
nothing@1:     ))
nothing@1:     
nothing@1:     if word[4] == 'V':
nothing@1:         temp = 'parameterized'
nothing@4:         dim = 'multi-dimensional'
nothing@1:     else:
nothing@1:         temp = word[4]
nothing@4:         if int(temp) == 1:
nothing@4:             dim = 'one-dimensional'
nothing@4:         else:
nothing@4:             dim = 'multi-dimensional'
nothing@1:         
nothing@1:     graph.add(( 
nothing@1:         URIRef(id),
nothing@1:         URIRef(local + 'dimensions'), 
nothing@1:         Literal(temp) 
nothing@1:     ))
nothing@4: 
nothing@4:     graph.add(( 
nothing@4:         URIRef(id),
nothing@4:         URIRef(local + 'dimensionality'), 
nothing@4:         Literal(dim) 
nothing@4:     ))
nothing@1:     
nothing@1:     temp = {
nothing@1:         'ASR': "speech recognition",
nothing@1:         'ESR': "environmental sound recognition",
nothing@1:         'MIR': "music information retrieval",
nothing@1:         'AS': "audio segmentation",
nothing@1:         'FP': "fingerprinting",
nothing@1:         'VAR': "several",
nothing@1:         'EXC': ''
nothing@1:     }[word[5]]
nothing@1: 
nothing@1:     if temp != '':   
nothing@1:         graph.add(( 
nothing@1:             URIRef(id), 
nothing@1:             URIRef(local + 'appdomain'), 
nothing@1:             Literal(temp) 
nothing@1:         ))
nothing@1:         
nothing@1:     steps = sig[i].split(' ')
nothing@1:     
nothing@1:     for key in steps:
nothing@1:         graph.add(( 
nothing@1:             URIRef(id), 
nothing@1:             URIRef(local + 'computation'), 
nothing@1:             Literal(compdict[key]) 
nothing@1:         ))
nothing@1:         
nothing@1:     if name.find('MPEG-7') >= 0:
nothing@1:         graph.add(( 
nothing@1:             URIRef(id),
nothing@1:             URIRef(local + 'computedIn'),
nothing@1:             Literal('MPEG-7')
nothing@1:         ))
nothing@1:     
nothing@1:     if name in abbr.keys():
nothing@1:         graph.add(( 
nothing@1:             URIRef(id),
nothing@1:             URIRef(local + 'abbreviation'),
nothing@1:             Literal(abbr[name])
nothing@1:         ))
nothing@1:         
nothing@1:     
nothing@1:     i += 1
nothing@1: 
nothing@1: 
nothing@1: graph.serialize('/Users/alo/MusicOntology/features/rdf/base.rdf')