def load_dataset(self, abox, tbox): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = (None, rdflib.RDF.type, rdflib.URIRef("http://purl.org/crmeh#EHE0009_ContextFind")) # define context # spoor with vulling context = [ rdflib.URIRef("http://purl.org/dc/elements/1.1/source"), [ rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P2F_has_type"), rdflib.RDF.value ], [ rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P45F_consists_of"), rdflib.RDF.value ] ] kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], context=context, strict_context=False) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = (None, rdflib.RDF.type, rdflib.URIRef("http://purl.org/crmeh#EHE1002_ContextFindProductionEvent")) # define context # spoor with vulling context = [rdflib.URIRef("http://purl.org/dc/elements/1.1/source"), (rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P4F_has_time-span"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P1F_is_identified_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P2F_has_type"), rdflib.RDF.value), (rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P108F_has_produced"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P2F_has_type"), rdflib.RDF.value)] kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], context=context, strict_context=False) return (kg_i_sampled, kg_s)
def cli(rule_base, abox=None, tbox=None, vocab=None, path="./of/rule_set", mode="uri", lang="en", overwrite=True, compress=False): logger.info("Initiating CLI") if abox is not None: logger.info("Loading ABox Graph") abox = rdf.read(local_path=abox) if tbox is not None: logger.info("Loading TBox Graph") tbox = rdf.read(local_path=tbox) if vocab is not None: logger.info("Loading Controlled Vocabulary") vocab = rdf.read(local_path=vocab) printer = None writer = None if mode == "label": printer = _rule_to_label_string writer = pretty_label_write elif mode == "natural": printer = _rule_to_natural_text writer = natural_write else: # mode == uri writer = pretty_write printer = _rule_to_string _ui(rule_base, abox, tbox, vocab, path, printer, writer, lang, overwrite, compress)
def load_dataset(self, abox, tbox): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) kg_i_sampled = KnowledgeGraph() for s, p, o in kg_i.triples(): if type(o) is rdflib.Literal: continue kg_i_sampled.graph.add((s, p, o)) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox, parameters): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = (None, None, rdflib.URIRef("http://purl.org/crmeh#EHE0007_Context")) kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], depth=parameters["sample_depth"]) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox, parameters): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = ( None, None, rdflib.URIRef("http://purl.org/collections/nl/dss/das/Voyage")) kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], depth=parameters["sample_depth"]) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) kg_i_sampled = KnowledgeGraph() for s, p, o in kg_i.triples(): if type(o) is rdflib.Resource: for ctype in kg_i_sampled.graph.objects(o, rdflib.type): if ctype == rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/E55_Type") or\ ctype == rdflib.URIRef("http://www.w3.org/2004/02/skos/core#Concept"): kg_i_sampled.graph.add((s, p, o)) break continue kg_i_sampled.graph.add((s, p, o)) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) return (kg_i, kg_s)
def load_dataset(self, abox, tbox): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = (None, rdflib.RDF.type, rdflib.URIRef("http://purl.org/crmeh#EHE0001_EHProject")) # define context # spoor with vulling context = [ rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_onderzoektype" ), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P7_took_place_at"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_gemeentecode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P7_took_place_at"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_plaatscode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P7_took_place_at"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_provinciecode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P7_took_place_at"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_toponiem" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P7_took_place_at"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_vindplaatstype" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P7_took_place_at"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P4_has_time-span"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_beginperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P7_took_place_at"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P4_has_time-span"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_eindperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P9_consists_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P108_has_produced"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_artefacttype" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P9_consists_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P108_has_produced"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_materiaalcategorie" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P9_consists_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P108_has_produced"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_verzamelwijze" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P9_consists_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P108_has_produced"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46_is_composed_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53_has_former_or_current_location" ), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_contexttype" )) ] kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], context=context, strict_context=False) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox): """ # pakbonLD SPARQL endpoint endpoint = "http://pakbon-ld.spider.d2s.labs.vu.nl/sparql/" # query query_string = "" " prefix pbont: <http://pakbon-ld.spider.d2s.labs.vu.nl/ont/> prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> SELECT DISTINCT ?s ?p ?o WHERE { ?s a pbont:SIKB0102S_Vondstcontext; ?p ?o. FILTER (?p != rdf:type) } LIMIT 1000"" " # perform query and return a KnowledgeGraph instance kg_i = rdf.query(query_string, endpoint) """ # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = ( None, rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_grondspoortype" ), None) # define context # spoor with vulling context = [ rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_grondspoortype" ), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P89_falls_within"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_contexttype" )), (rdflib.URIRef("http://purl.org/crmeh#EHP3i"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_kleur") ), (rdflib.URIRef("http://purl.org/crmeh#EHP3i"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_textuur" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_structuurtype" )), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_diepte"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P40_observed_dimension"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P90_has_value") ), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_diepte"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P40_observed_dimension"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P91_has_unit")), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_beginperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_eindperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_beginperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_eindperiode" )) ] kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], context=context) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox): """ # pakbonLD SPARQL endpoint endpoint = "http://pakbon-ld.spider.d2s.labs.vu.nl/sparql/" # query query_string = "" " prefix pbont: <http://pakbon-ld.spider.d2s.labs.vu.nl/ont/> prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> SELECT DISTINCT ?s ?p ?o WHERE { ?s a pbont:SIKB0102S_Vondstcontext; ?p ?o. FILTER (?p != rdf:type) } LIMIT 1000"" " # perform query and return a KnowledgeGraph instance kg_i = rdf.query(query_string, endpoint) """ # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = (None, rdflib.RDF.type, rdflib.URIRef("http://purl.org/crmeh#EHE0009_ContextFind")) # define context # spoor with vulling context = [ rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_artefacttype" ), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_exposabel" ), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_geconserveerd" ), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_gedeselecteerd" ), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_materiaalcategorie" ), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P3_has_note"), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_gewicht" ), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P90_has_value") ), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_gewicht" ), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P91_has_unit")), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_aantal"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P90_has_value") ), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_aantal"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P91_has_unit")), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_verzamelwijze" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_beginperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_eindperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_bewaarTemperatuur" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_bewaarVochtigheid" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_breekbaar" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P46i_forms_part_of"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_lichtgevoelig" )) ] kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], context=context, strict_context=False) return (kg_i_sampled, kg_s)
def load_dataset(self, abox, tbox): # read graphs kg_i = rdf.read(local_path=abox) kg_s = rdf.read(local_path=tbox) # sample by pattern pattern = ( None, rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_grondspoortype" ), None) # define context # spoor with vulling context = [ rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_grondspoortype" ), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P89_falls_within"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_contexttype" )), (rdflib.URIRef("http://purl.org/crmeh#EHP3i"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_kleur") ), (rdflib.URIRef("http://purl.org/crmeh#EHP3i"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_textuur" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_structuurtype" )), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_diepte"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P40_observed_dimension"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P90_has_value") ), (rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_diepte"), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P40_observed_dimension"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P91_has_unit")), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_beginperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_eindperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_beginperiode" )), (rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P53i_is_former_or_current_location_of" ), rdflib.URIRef( "http://www.cidoc-crm.org/cidoc-crm/P140i_was_attributed_by"), rdflib.URIRef("http://www.cidoc-crm.org/cidoc-crm/P141_assigned"), rdflib.URIRef( "http://pakbon-ld.spider.d2s.labs.vu.nl/ont/SIKB0102S_eindperiode" )) ] kg_i_sampled = kg_i.sample(sampler, patterns=[pattern], context=context, strict_context=True) return (kg_i_sampled, kg_s)