class TestDescriber(unittest.TestCase):
def setUp(self):
self.graph = Graph()
self.graph.parse(data="""
@prefix dcterms: <http://purl.org/dc/terms/> .
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
<http://example.org/doc> a foaf:Document;
dcterms:title "Hello world"@en ;
dcterms:identifier "ID1",
"ID2";
dcterms:issued "2013-10-11"^^xsd:date;
dcterms:references <http://example.org/doc2>;
dcterms:subject <http://example.org/concept1>,
<http://example.org/concept2> .
""", format="turtle")
self.desc = Describer(self.graph, "http://example.org/doc")
def test_getvalues(self):
self.assertEqual(self.desc.getvalues(DCTERMS.alternate),
[])
self.assertEqual(self.desc.getvalues(DCTERMS.title),
["Hello world"])
self.assertEqual(set(self.desc.getvalues(DCTERMS.identifier)),
set(["ID1", "ID2"]))
def test_getvalue(self):
self.assertEqual(self.desc.getvalue(DCTERMS.title),
"Hello world")
self.assertEqual(self.desc.getvalue(DCTERMS.issued),
datetime.date(2013,10,11))
with self.assertRaises(KeyError):
self.desc.getvalue(DCTERMS.alternate)
with self.assertRaises(KeyError):
self.desc.getvalue(DCTERMS.identifier)
def test_getrels(self):
self.assertEqual(self.desc.getrels(DCTERMS.replaces),
[])
self.assertEqual(self.desc.getrels(DCTERMS.references),
["http://example.org/doc2"])
self.assertEqual(set(self.desc.getrels(DCTERMS.subject)),
set(["http://example.org/concept1",
"http://example.org/concept2"]))
def test_getrel(self):
self.assertEqual(self.desc.getrel(DCTERMS.references),
"http://example.org/doc2")
with self.assertRaises(KeyError):
self.desc.getrel(DCTERMS.replaces)
with self.assertRaises(KeyError):
self.desc.getrel(DCTERMS.subject)
def test_getrdftype(self):
self.assertEqual(self.desc.getrdftype(),
"http://xmlns.com/foaf/0.1/Document")
class TestDescriber(unittest.TestCase):
def setUp(self):
self.graph = Graph()
self.graph.parse(data="""
@prefix dcterms: <http://purl.org/dc/terms/> .
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
<http://example.org/doc> a foaf:Document;
dcterms:title "Hello world"@en ;
dcterms:identifier "ID1",
"ID2";
dcterms:issued "2013-10-11"^^xsd:date;
dcterms:references <http://example.org/doc2>;
dcterms:subject <http://example.org/concept1>,
<http://example.org/concept2> .
""",
format="turtle")
self.desc = Describer(self.graph, "http://example.org/doc")
def test_getvalues(self):
self.assertEqual(self.desc.getvalues(DCTERMS.alternate), [])
self.assertEqual(self.desc.getvalues(DCTERMS.title), ["Hello world"])
self.assertEqual(set(self.desc.getvalues(DCTERMS.identifier)),
set(["ID1", "ID2"]))
def test_getvalue(self):
self.assertEqual(self.desc.getvalue(DCTERMS.title), "Hello world")
self.assertEqual(self.desc.getvalue(DCTERMS.issued),
datetime.date(2013, 10, 11))
with self.assertRaises(KeyError):
self.desc.getvalue(DCTERMS.alternate)
with self.assertRaises(KeyError):
self.desc.getvalue(DCTERMS.identifier)
def test_getrels(self):
self.assertEqual(self.desc.getrels(DCTERMS.replaces), [])
self.assertEqual(self.desc.getrels(DCTERMS.references),
["http://example.org/doc2"])
self.assertEqual(
set(self.desc.getrels(DCTERMS.subject)),
set(["http://example.org/concept1",
"http://example.org/concept2"]))
def test_getrel(self):
self.assertEqual(self.desc.getrel(DCTERMS.references),
"http://example.org/doc2")
with self.assertRaises(KeyError):
self.desc.getrel(DCTERMS.replaces)
with self.assertRaises(KeyError):
self.desc.getrel(DCTERMS.subject)
def test_getrdftype(self):
self.assertEqual(self.desc.getrdftype(),
"http://xmlns.com/foaf/0.1/Document")
def infer_metadata(self, resource, basefile):
# remove the bogus dcterms:issued thing that we only added to
# aid URI generation. NB: This is removed in the superclass'
# postprocess_doc as well, because for this lagen.nu-derived
# class it needs to be done at this point, but for use of the
# superclass directly, it needs to be done at some point.
for o in resource.objects(DCTERMS.issued):
if not o.datatype:
resource.remove(DCTERMS.issued, o)
sameas_uri = self.sameas_minter.space.coin_uri(resource)
resource.add(OWL.sameAs, URIRef(sameas_uri))
resource.graph.add((URIRef(self.canonical_uri(basefile, True)),
OWL.sameAs, resource.identifier))
# then find each rpubl:konsolideringsunderlag, and create
# owl:sameas for them as well
for subresource in resource.objects(RPUBL.konsolideringsunderlag):
# sometimes there'll be a rpubl:konsolideringsunderlag to
# a resource URI but no actual data about that
# resource. This seems to happen if SFST is updated but
# SFSR is not. In those cases we can't generate a
# owl:sameAs URI since we have no other data about the
# resource.
if subresource.value(RDF.type):
uri = self.sameas_minter.space.coin_uri(subresource)
subresource.add(OWL.sameAs, URIRef(uri))
desc = Describer(resource.graph, resource.identifier)
de = DocumentEntry(self.store.documententry_path(basefile))
if de.orig_updated:
desc.value(RINFOEX.senastHamtad, de.orig_updated)
if de.orig_checked:
desc.value(RINFOEX.senastKontrollerad, de.orig_checked)
rooturi = URIRef(desc.getrel(RPUBL.konsoliderar))
v = self.commondata.value(rooturi, DCTERMS.alternate, any=True)
if v:
desc.value(DCTERMS.alternate, v)
v = self.commondata.value(rooturi, RDFS.label, any=True)
if v:
# don't include labels if they're essentially the same as
# dcterms:title (legalref needs it to be able to parse
# refs to laws that typically don't include SFS numbers,
# so that's why they're in sfs.ttl
basetitle = str(resource.value(DCTERMS.title)).rsplit(" (")[0]
if not v.startswith(basetitle.lower()):
desc.value(RDFS.label, util.ucfirst(v))
def infer_metadata(self, resource, basefile):
# remove the bogus dcterms:issued thing that we only added to
# aid URI generation. NB: This is removed in the superclass'
# postprocess_doc as well, because for this lagen.nu-derived
# class it needs to be done at this point, but for use of the
# superclass directly, it needs to be done at some point.
for o in resource.objects(DCTERMS.issued):
if not o.datatype:
resource.remove(DCTERMS.issued, o)
sameas_uri = self.sameas_minter.space.coin_uri(resource)
resource.add(OWL.sameAs, URIRef(sameas_uri))
resource.graph.add((URIRef(self.canonical_uri(basefile, True)),
OWL.sameAs, resource.identifier))
# then find each rpubl:konsolideringsunderlag, and create
# owl:sameas for them as well
for subresource in resource.objects(RPUBL.konsolideringsunderlag):
# sometimes there'll be a rpubl:konsolideringsunderlag to
# a resource URI but no actual data about that
# resource. This seems to happen if SFST is updated but
# SFSR is not. In those cases we can't generate a
# owl:sameAs URI since we have no other data about the
# resource.
if subresource.value(RDF.type):
uri = self.sameas_minter.space.coin_uri(subresource)
subresource.add(OWL.sameAs, URIRef(uri))
desc = Describer(resource.graph, resource.identifier)
de = DocumentEntry(self.store.documententry_path(basefile))
if de.orig_updated:
desc.value(RINFOEX.senastHamtad, de.orig_updated)
if de.orig_checked:
desc.value(RINFOEX.senastKontrollerad, de.orig_checked)
rooturi = URIRef(desc.getrel(RPUBL.konsoliderar))
v = self.commondata.value(rooturi, DCTERMS.alternate, any=True)
if v:
desc.value(DCTERMS.alternate, v)
v = self.commondata.value(rooturi, RDFS.label, any=True)
if v:
# don't include labels if they're essentially the same as
# dcterms:title (legalref needs it to be able to parse
# refs to laws that typically don't include SFS numbers,
# so that's why they're in sfs.ttl
basetitle = str(resource.value(DCTERMS.title)).rsplit(" (")[0]
if not v.startswith(basetitle.lower()):
desc.value(RDFS.label, util.ucfirst(v))
def selector(entry):
graph = Graph()
with self.store.open_distilled(entry.basefile) as fp:
graph.parse(data=fp.read())
desc = Describer(graph, entry.id)
return desc.getrel(self.ns['dcterms'].subject) == category
def selector(entry):
graph = Graph()
with self.store.open_distilled(entry.basefile) as fp:
graph.parse(data=fp.read())
desc = Describer(graph, entry.id)
return desc.getrel(self.ns['dcterms'].subject) == category
