class TestPrefLabel(unittest.TestCase):
def setUp(self):
self.g = ConjunctiveGraph()
self.u = URIRef('http://example.com/foo')
self.g.add([self.u, RDFS.label, Literal('foo')])
self.g.add([self.u, RDFS.label, Literal('bar')])
def test_default_label_sorting(self):
res = sorted(self.g.preferredLabel(self.u))
tgt = [(rdflib.term.URIRef('http://www.w3.org/2000/01/rdf-schema#label'),
rdflib.term.Literal(u'bar')),
(rdflib.term.URIRef('http://www.w3.org/2000/01/rdf-schema#label'),
rdflib.term.Literal(u'foo'))]
self.assertEqual(res, tgt)
def test_default_preflabel_sorting(self):
self.g.add([self.u, SKOS.prefLabel, Literal('bla')])
res = self.g.preferredLabel(self.u)
tgt = [(rdflib.term.URIRef('http://www.w3.org/2004/02/skos/core#prefLabel'),
rdflib.term.Literal(u'bla'))]
self.assertEqual(res, tgt)
def test_preflabel_lang_sorting_no_lang_attr(self):
self.g.add([self.u, SKOS.prefLabel, Literal('bla')])
self.g.add([self.u, SKOS.prefLabel, Literal('blubb', lang='en')])
res = sorted(self.g.preferredLabel(self.u))
tgt = [(rdflib.term.URIRef('http://www.w3.org/2004/02/skos/core#prefLabel'),
rdflib.term.Literal(u'bla')),
(rdflib.term.URIRef('http://www.w3.org/2004/02/skos/core#prefLabel'),
rdflib.term.Literal(u'blubb', lang='en'))]
self.assertEqual(res, tgt)
def test_preflabel_lang_sorting_empty_lang_attr(self):
self.g.add([self.u, SKOS.prefLabel, Literal('bla')])
self.g.add([self.u, SKOS.prefLabel, Literal('blubb', lang='en')])
res = self.g.preferredLabel(self.u, lang='')
tgt = [(rdflib.term.URIRef('http://www.w3.org/2004/02/skos/core#prefLabel'),
rdflib.term.Literal(u'bla'))]
self.assertEqual(res, tgt)
def test_preflabel_lang_sorting_en_lang_attr(self):
self.g.add([self.u, SKOS.prefLabel, Literal('blubb', lang='en')])
res = self.g.preferredLabel(self.u, lang='en')
tgt = [(rdflib.term.URIRef('http://www.w3.org/2004/02/skos/core#prefLabel'),
rdflib.term.Literal(u'blubb', lang='en'))]
self.assertEqual(res, tgt)
from rdflib import ConjunctiveGraph, URIRef, RDFS, Literal
from rdflib.namespace import SKOS
from pprint import pprint
g = ConjunctiveGraph()
u = URIRef(u'http://example.com/foo')
g.add([u, RDFS.label, Literal('foo')]) # 给u添加标签
g.add([u, RDFS.label, Literal('bar')]) # 给u添加标签
pprint(sorted(g.preferredLabel(u)))
g.add([u, SKOS.prefLabel, Literal('bla')])
pprint(g.preferredLabel(u))
g.add([u, SKOS.prefLabel, Literal('blubb', lang='en')])
sorted(g.preferredLabel(u))
g.preferredLabel(u, lang='')
pprint(g.preferredLabel(u, lang='en'))
"alt_labels_es": {"type": "text", "analyzer": "autocomplete", "search_analyzer": "autocomplete_search"},
}
}
print("creating mapping ...")
res = es_con.indices.put_mapping(index=index_name, doc_type="doc", body=thesaurus_mapping)
print("resonse: {}".format(res))
i = 0
for uri in graph.query(querystring):
this_uri = uri[0]
doc = {"uri": this_uri}
j = 0
for lang in ['ar', 'zh', 'en', 'fr', 'ru', 'es']:
pref_labels = []
for label in graph.preferredLabel(URIRef(this_uri), lang):
pref_labels.append(label[1])
doc.update({"labels_{}".format(lang): pref_labels})
alt_labels = []
for label in graph.objects(URIRef(this_uri), SKOS.altLabel):
if label.language == lang:
alt_labels.append(label)
doc.update({"alt_labels_{}".format(lang): alt_labels})
payload = json.dumps(doc)
res = es_con.index(index=index_name, doc_type='doc', body=payload)
doc = {"uri": this_uri}
j += 1
i += j
class TestPrefLabel(unittest.TestCase):
def setUp(self):
self.g = ConjunctiveGraph()
self.u = URIRef("https://example.com/foo")
self.g.add((self.u, RDFS.label, Literal("foo")))
self.g.add((self.u, RDFS.label, Literal("bar")))
def test_default_label_sorting(self):
res = sorted(self.g.preferredLabel(self.u))
tgt = [
(
rdflib.term.URIRef("http://www.w3.org/2000/01/rdf-schema#label"),
rdflib.term.Literal("bar"),
),
(
rdflib.term.URIRef("http://www.w3.org/2000/01/rdf-schema#label"),
rdflib.term.Literal("foo"),
),
]
self.assertEqual(res, tgt)
def test_default_preflabel_sorting(self):
self.g.add((self.u, SKOS.prefLabel, Literal("bla")))
res = self.g.preferredLabel(self.u)
tgt = [
(
rdflib.term.URIRef("http://www.w3.org/2004/02/skos/core#prefLabel"),
rdflib.term.Literal("bla"),
)
]
self.assertEqual(res, tgt)
def test_preflabel_lang_sorting_no_lang_attr(self):
self.g.add((self.u, SKOS.prefLabel, Literal("bla")))
self.g.add((self.u, SKOS.prefLabel, Literal("blubb", lang="en")))
res = sorted(self.g.preferredLabel(self.u))
tgt = [
(
rdflib.term.URIRef("http://www.w3.org/2004/02/skos/core#prefLabel"),
rdflib.term.Literal("bla"),
),
(
rdflib.term.URIRef("http://www.w3.org/2004/02/skos/core#prefLabel"),
rdflib.term.Literal("blubb", lang="en"),
),
]
self.assertEqual(res, tgt)
def test_preflabel_lang_sorting_empty_lang_attr(self):
self.g.add((self.u, SKOS.prefLabel, Literal("bla")))
self.g.add((self.u, SKOS.prefLabel, Literal("blubb", lang="en")))
res = self.g.preferredLabel(self.u, lang="")
tgt = [
(
rdflib.term.URIRef("http://www.w3.org/2004/02/skos/core#prefLabel"),
rdflib.term.Literal("bla"),
)
]
self.assertEqual(res, tgt)
def test_preflabel_lang_sorting_en_lang_attr(self):
self.g.add((self.u, SKOS.prefLabel, Literal("blubb", lang="en")))
res = self.g.preferredLabel(self.u, lang="en")
tgt = [
(
rdflib.term.URIRef("http://www.w3.org/2004/02/skos/core#prefLabel"),
rdflib.term.Literal("blubb", lang="en"),
)
]
self.assertEqual(res, tgt)
def handle(self, *args, **options):
es = settings.ELASTIC_SEARCH_URL
db = os.path.join(settings.BASE_DIR, "db")
print(es)
graph = ConjunctiveGraph('Sleepycat')
graph.open(db, create=False)
graph.bind('skos', SKOS)
EU = Namespace('http://eurovoc.europa.eu/schema#')
UNBIST = Namespace('http://unontologies.s3-website-us-east-1.amazonaws.com/unbist#')
querystring = "select ?uri where { ?uri rdf:type skos:Concept filter not exists { ?uri rdf:type unbist:PlaceName } . }"
index = 1
# make the index:
thes_index = {
"mappings": {
"terms": {
"properties": {
"scope_notes": {
"type": "string"
},
"uri": {
"type": "string"
},
"alt_labels": {
"type": "string"
},
"alt_labels_orig": {
"type": "string",
"index": "not_analyzed"
},
"labels": {
"type": "string"
},
"labels_orig": {
"type": "string",
"index": "not_analyzed"
}
}
}
}
}
r = requests.put(es + 'thesaurus/', data=json.dumps(thes_index))
for uri in graph.query(querystring):
this_uri = uri[0]
doc = {
"uri": this_uri
}
pref_labels = []
labels_orig_lc = []
print("Getting preferred labels")
for label in graph.preferredLabel(URIRef(this_uri)):
pref_labels.append(label[1])
if label[1].language in ['en','fr','es']:
labels_orig_lc.append(label[1].lower())
doc.update({"labels": pref_labels})
doc.update({"labels_orig": pref_labels + labels_orig_lc})
alt_labels = []
alt_labels_orig_lc = []
print("Getting alternate labels")
for label in graph.objects(URIRef(this_uri), SKOS.altLabel):
alt_labels.append(label)
if label.language in ['en','fr','es']:
alt_labels_orig_lc.append(label.lower())
doc.update({"alt_labels": alt_labels})
doc.update({"alt_labels_orig": alt_labels + alt_labels_orig_lc})
scope_notes = []
print("Getting scope notes")
for sn in graph.objects(URIRef(this_uri), SKOS.scopeNote):
scope_notes.append(sn)
doc.update({"scope_notes": scope_notes})
payload = json.dumps(doc)
r = requests.put(es + 'thesaurus/terms/' + str(index), data=payload)
index += 1
