def test_creator_names_picks_up_contributors(self):
graph = json.loads(
self.sample_data("no_author_only_contributor.jsonld"))['@graph']
eq_(([], []), OCLCLinkedData.creator_names(graph))
eq_((['Thug Kitchen LLC.'], []),
OCLCLinkedData.creator_names(graph, 'contributor'))
def test_creator_names_gathers_external_uris(self):
graph = json.loads(
self.sample_data("creator_includes_viaf_uris.jsonld"))['@graph']
names, uris = OCLCLinkedData.creator_names(graph)
eq_([], names)
eq_(set(["http://id.loc.gov/authorities/names/n2013058227",
"http://viaf.org/viaf/221233754",
"http://viaf.org/viaf/305306689"]),
set(uris))
def test_book_info_to_metadata(self):
oclc = OCLCLinkedData(self._db)
subgraph = json.loads(self.sample_data("galapagos.jsonld"))['@graph']
[book] = [book for book in oclc.books(subgraph)]
metadata_obj = OCLCLinkedData(self._db).book_info_to_metadata(
subgraph, book
)
# A metadata object is returned, with the proper OCLC identifier.
eq_(True, isinstance(metadata_obj, Metadata))
eq_(Identifier.OCLC_NUMBER, metadata_obj.primary_identifier.type)
eq_(u"11866009", metadata_obj.primary_identifier.identifier)
# It has publication information & ISBNs
eq_(u"Galápagos : a novel", metadata_obj.title)
eq_(u'Delacorte Press/Seymour Lawrence', metadata_obj.publisher)
eq_(1985, metadata_obj.published.year)
eq_(1, len(metadata_obj.links))
assert "ghost of a shipbuilder" in metadata_obj.links[0].content
eq_(4, len(metadata_obj.identifiers))
eq_(1, len(metadata_obj.contributors))
[viaf] = [c.viaf for c in metadata_obj.contributors]
eq_(u"71398958", viaf)
eq_(10, len(metadata_obj.subjects))
# Make sure a book with no English title doesn't break anything.
subgraph[14]['name']['@language'] = 'fr'
[book] = [book for book in oclc.books(subgraph)]
metadata_obj = OCLCLinkedData(self._db).book_info_to_metadata(
subgraph, book
)
# The metadata has no title.
eq_(None, metadata_obj.title)
def test_extract_useful_data(self):
subgraph = json.loads(
self.sample_data('galapagos.jsonld')
)['@graph']
[book] = [book for book in OCLCLinkedData.books(subgraph)]
(oclc_id_type,
oclc_id,
titles,
descriptions,
subjects,
creator_uris,
publishers,
publication_dates,
example_uris) = OCLCLinkedData.extract_useful_data(subgraph, book)
eq_(Identifier.OCLC_NUMBER, oclc_id_type)
eq_(u"11866009", oclc_id)
eq_([u"Galápagos : a novel"], titles)
eq_(1, len(descriptions))
# Even though there are 11 links in the books "about" list,
# "http://subject.example.wo/internal_lookup" does not get included as
# a subject because it doesn't have an internal lookup.
eq_(1, len(subjects[Subject.DDC]))
eq_(1, len(subjects[Subject.FAST]))
eq_(4, len(subjects[Subject.TAG]))
eq_(1, len(subjects[Subject.PLACE]))
# Meanwhile, the made-up LCSH subject that also doesn't have an
# internal lookup is included because its details can be parsed from
# the url: "http://id.loc.gov/authorities/subjects/sh12345678"
eq_(3, len(subjects[Subject.LCSH]))
eq_(1, len(creator_uris))
eq_(["Delacorte Press/Seymour Lawrence"], publishers)
eq_(["1985"], publication_dates)
eq_(2, len(example_uris))
def test_extract_contributor(self):
# It pulls relevant contributor data out of an OCLC person entity graph.
sloane_info = json.loads(
self.sample_data('sloane_crosley.jsonld'))['@graph'][1]
result = OCLCLinkedData.extract_contributor(sloane_info)
eq_(result['family_name'], 'Crosley')
eq_(result['display_name'], 'Sloane Crosley')
flanagan_info = json.loads(
self.sample_data('john_flanagan_multiname.jsonld'))
flanagan_info = flanagan_info['@graph'][1]
result = OCLCLinkedData.extract_contributor(flanagan_info)
eq_(result['family_name'], 'Flanagan')
eq_(result['display_name'], 'John Anthony Flanagan')
eq_(result['extra']['birthDate'], '1944')
# TODO: Modify the contributor extraction to handle cases where
# maiden names are included and/or multiple name options are the
# same except for capitalization.
rice_info = json.loads(self.sample_data('anne_rice.jsonld'))
result = OCLCLinkedData.extract_contributor(rice_info['@graph'][1])
eq_(result['family_name'], "O'Brien Rice")
eq_(result['display_name'], "Anne O'Brien Rice")
eq_(result['extra']['birthDate'], '1941')
def __init__(self, _db, oclcld=None, viaf=None):
self._db = _db
self.oclcld = oclcld or OCLCLinkedData(_db)
self.viaf = viaf or VIAFClient(_db)
self.log = logging.getLogger("Author name canonicalizer")
