def test_add_doc_data_csv_base(conn, ex, uri):
conn.addDocumentData('a,b,c\ndd,ee,ff', DocFormat.CSV, base=uri)
statements = get_statements(conn)
assert 3 == len(statements)
subject = statements[0][0]
assert statements == [[subject, ex.a, Literal('dd'), None],
[subject, ex.b, Literal('ee'), None],
[subject, ex.c, Literal('ff'), None]]
def test_add_doc_data_context(conn, ex, uri):
conn.addDocumentData({'a': 'aa', 'b': 'bb'}, context=uri)
statements = get_statements(conn)
assert 2 == len(statements)
subject = statements[0][0]
ns = conn.namespace(statements[0][1].getNamespace())
assert statements == [[subject, ns.a, Literal('aa'), ex.g],
[subject, ns.b, Literal('bb'), ex.g]]
def test_add_doc_data_subject(conn, ex, uri):
conn.addDocumentData('a,b,c\ndd,ee,ff', DocFormat.CSV, subject=uri)
statements = get_statements(conn)
assert 3 == len(statements)
ns = conn.namespace(statements[0][1].getNamespace())
assert statements == [[ex.s, ns.a, Literal('dd'), None],
[ex.s, ns.b, Literal('ee'), None],
[ex.s, ns.c, Literal('ff'), None]]
def test_add_doc_data_csv_escape(conn):
conn.addDocumentData('a,b,c\nxx,"y?"y",zz', DocFormat.CSV, csv_escape='?')
statements = get_statements(conn)
assert 3 == len(statements)
subject = statements[0][0]
ns = conn.namespace(statements[0][1].getNamespace())
assert statements == [[subject, ns.a, Literal('xx'), None],
[subject, ns.b, Literal('y"y'), None],
[subject, ns.c, Literal('zz'), None]]
def test_add_doc_data_csv_separator(conn):
conn.addDocumentData('a:b:c\nxx:yy:zz', DocFormat.CSV, csv_separator=':')
statements = get_statements(conn)
assert 3 == len(statements)
subject = statements[0][0]
ns = conn.namespace(statements[0][1].getNamespace())
assert statements == [[subject, ns.a, Literal('xx'), None],
[subject, ns.b, Literal('yy'), None],
[subject, ns.c, Literal('zz'), None]]
def test_add_doc_data_csv(conn):
conn.addDocumentData('a,b,c\ndd,ee,ff', DocFormat.CSV)
statements = get_statements(conn)
assert 3 == len(statements)
subject = statements[0][0]
ns = conn.namespace(statements[0][1].getNamespace())
assert statements == [[subject, ns.a, Literal('dd'), None],
[subject, ns.b, Literal('ee'), None],
[subject, ns.c, Literal('ff'), None]]
def test_add_json_ld_integer_literal_value(conn, example):
conn.addData({
'@context': {
'@vocab': 'http://franz.com/example/',
'@base': 'http://franz.com/example/'
},
'@id': 's',
example.p: Literal(42)
})
assert [[example.s, example.p, Literal(42), None]] == get_statements(conn)
def test_add_doc_data_csv_base_and_subject(conn, ex):
conn.addDocumentData('a,b,c\ndd,ee,ff',
DocFormat.CSV,
base='ex://',
subject='ex://$a')
assert get_statements(conn) == [[ex.dd, ex.a,
Literal('dd'), None],
[ex.dd, ex.b,
Literal('ee'), None],
[ex.dd, ex.c,
Literal('ff'), None]]
def test_add_doc_data_csv_dialect(conn, excel_tab):
conn.addDocumentData('a\tb\tc\r\n"x""x"\t yy\tzz',
DocFormat.CSV,
csv_dialect=excel_tab)
statements = get_statements(conn)
assert 3 == len(statements)
subject = statements[0][0]
ns = conn.namespace(statements[0][1].getNamespace())
assert statements == [[subject, ns.a, Literal('x"x'), None],
[subject, ns.b, Literal(' yy'), None],
[subject, ns.c, Literal('zz'), None]]
def test_add_doc_data_csv_override_dialect(conn, doc, arg, value, dialect):
conn.addDocumentData(doc,
DocFormat.CSV,
csv_dialect=dialect,
**{arg: value})
statements = get_statements(conn)
assert 3 == len(statements)
subject = statements[0][0]
ns = conn.namespace(statements[0][1].getNamespace())
assert statements == [[subject, ns.a, Literal('x"x'), None],
[subject, ns.b, Literal(' yy'), None],
[subject, ns.c, Literal('zz'), None]]
def test_add_json_ld_typed_literal_value(conn, example):
conn.addData({
'@context': {
'@vocab': 'http://franz.com/example/',
'@base': 'http://franz.com/example/'
},
'@id': 's',
example.p: Literal('oooo', XMLSchema.BASE64BINARY)
})
assert [[
example.s, example.p,
Literal('oooo', XMLSchema.BASE64BINARY), None
]] == get_statements(conn)
def test_add_json_ld_lang_literal_value(conn, example):
conn.addData({
'@context': {
'@vocab': 'http://franz.com/example/',
'@base': 'http://franz.com/example/'
},
'@id':
's',
# A bathtub in Sindarin
example.p:
Literal('🛀', language='sjn')
})
assert [[example.s, example.p,
Literal('🛀', language='sjn'), None]] == get_statements(conn)
def test_add_doc_data_prefix_in_keys(conn, pref1, pref2):
p1 = conn.namespace('p1://')
p2 = conn.namespace('p2://')
conn.addDocumentData({
'a': 'aa',
'b': 'bb'
},
keys={
'a': DocumentKey(prefix=pref1),
'b': DocumentKey(prefix=pref2)
})
statements = get_statements(conn)
assert len(statements) == 2
assert statements == [[statements[0][0], p1.a,
Literal('aa'), None],
[statements[1][0], p2.b,
Literal('bb'), None]]
def test_add_doc_data_attributes(conn, ex, attr):
conn.addDocumentData({'p': 'oo'},
base=ex(''),
subject=ex.s,
attributes={'test': 'c'})
assert [{
'test': 'c'
}] == get_triple_attributes(conn, ex.s, ex.p, Literal('oo'))
def test_add_doc_data_store_source(conn):
# Empty dict will not be stored at all, so we need at least one key
src = '{"a":"bb"}'
conn.addDocumentData(src, json_store_source=True)
for _s, _p, o, _g in get_statements(conn):
if o == Literal(src):
break
else:
pytest.fail('Source document not found.')
def test_add_doc_data_prefix_override(conn, pref1, pref2):
p2 = conn.namespace('p2://')
conn.addDocumentData({'a': 'aa'},
keys={
'a': DocumentKey(prefix=pref1),
},
prefix={'a': pref2})
statements = get_statements(conn)
assert len(statements) == 1
assert statements == [[statements[0][0], p2.a, Literal('aa'), None]]
def test_add_doc_data_rules(conn, ex):
conn.addData('''
@prefix ldm: <http://franz.com/ns/allegrograph/6.4/load-meta#> .
ldm:myrules ldm:id "ex://test${x}" ;
ldm:prefix "ex://" .''')
conn.addDocumentData({'x': "abc"}, rules='myrules')
statements = [
stmt for stmt in get_statements(conn)
if 'load-meta' not in str(stmt[0])
]
assert statements == [[ex.testabc, ex.x, Literal('abc'), None]]
def test_long_literal():
# This behavior of using LONG as the datatype might be convenient in demos,
# (you can do ``Literal(42L)``), but has a few drawbacks:
# - is incorrect: will map arbitrarily huge values to LONG, which is
# supposed to be 64 bit, while SMALLER values will become INTEGERS
# (which have no size limit).
# - In fact longs start at sys.maxint + 1 == 2 ** 63, i.e. as soon as
# it is no longer valid to hold them in xsd:longs.
# - Python 3 has no 'long' type.
literal = Literal(84104105115032109097107101115032110111032115101110115101046)
assert literal.datatype is XMLSchema.LONG
def test_add_doc_content_encoding(conn, after):
f = tempfile.NamedTemporaryFile(suffix='.json.gz', delete=False)
filename = f.name
after.append(lambda: os.remove(filename))
with gzip.GzipFile('data.json', 'wb', fileobj=f) as gz:
gz.write('{"p":"oo"}'.encode('utf-8'))
f.close()
conn.addDocumentFile(filename, DocFormat.JSON, content_encoding='gzip')
statements = get_statements(conn)
assert 1 == len(statements)
assert statements[0][2] == Literal("oo")
def test_add_doc_encoding(conn, after):
f = tempfile.NamedTemporaryFile(suffix='.csv', delete=False)
filename = f.name
after.append(lambda: os.remove(filename))
f.write(u'p\nрджреБрдкЁЯЫБ'.encode('utf-16le'))
f.close()
# This is completely broken... it expects an external format name here
conn.addDocumentFile(filename, DocFormat.CSV, encoding='unicode')
statements = get_statements(conn)
assert 1 == len(statements)
assert statements[0][2] == Literal(u"рджреБрдкЁЯЫБ")
def test_add_doc_data_transform_template(conn):
conn.addDocumentData({'a': 'XY'}, transform={'a': 'test-$a'})
statements = get_statements(conn)
assert len(statements) == 1
assert statements[0][2] == Literal('test-XY')
def test_literals_from_python_values(value, expected_text, expected_type):
literal = Literal(value)
assert literal.label == expected_text
# Well-known types are normalized, so it is safe to use the ``is`` operator here.
assert literal.datatype is expected_type
def test_true_literal():
assert Literal(True).booleanValue()
def test_false_literal():
assert not Literal(False).booleanValue()
def test_time_value_tz():
lit = Literal('1984-08-26T10:00:05+02:00', XMLSchema.DATETIME)
t = lit.datetimeValue()
assert t.utcoffset() == timedelta(0, 7200)
def test_add_doc_data_transform_built_in(conn):
conn.addDocumentData({'a': 'aa'}, transform={'a': 'string-capitalize'})
statements = get_statements(conn)
assert len(statements) == 1
assert statements[0][2] == Literal('Aa')
