def test_create():
s = fake.session()
dataset = fake.dataset()
attrs = tuple(
[
tc.SubAttribute(
name=str(i),
is_nullable=True,
type=tc.attribute.type.Array(tc.attribute.type.STRING),
)
for i in range(4)
]
)
attr = tc.attribute.create(
s,
dataset,
name="attr",
is_nullable=False,
type=tc.attribute.type.Record(attributes=attrs),
)
assert attr.name == "attr"
assert not attr.is_nullable
assert isinstance(attr.type, tc.attribute.type.Record)
assert attr.type.attributes == attrs
def test_upsert_infer_primary_key():
s = fake.session()
dataset = fake.dataset()
df = pd.DataFrame(_records_json)
response = tc.dataframe.upsert(s, dataset, df)
assert response == _response_json
def test_delete_primary_key_not_found():
s = fake.session()
dataset = fake.dataset()
with pytest.raises(tc.primary_key.NotFound):
tc.record.delete(s,
dataset,
_records_json,
primary_key_name="wrong_primary_key")
def test_delete():
s = fake.session()
dataset = fake.dataset()
response = tc.record.delete(s,
dataset,
_records_json,
primary_key_name="primary_key")
assert response == _response_json
def test_upsert_primary_key_not_found():
s = fake.session()
dataset = fake.dataset()
df = pd.DataFrame(_records_json)
with pytest.raises(tc.primary_key.NotFound):
tc.dataframe.upsert(s,
dataset,
df,
primary_key_name="wrong_primary_key")
def test_update():
s = fake.session()
dataset = fake.dataset()
updates = [
tc.record._create_command(record, primary_key_name="primary_key")
for record in _records_json
]
response = tc.record._update(s, dataset, updates)
assert response == _response_json
def test_upsert_index_column_name_collision():
s = fake.session()
dataset = fake.dataset()
df = pd.DataFrame(_records_json_2)
df.index.name = "primary_key"
# create column in `df` with same name as index and matching "primary_key"
df.insert(0, df.index.name, df.index)
with pytest.raises(tc.primary_key.Ambiguous):
tc.dataframe.upsert(s, dataset, df, primary_key_name="primary_key")
def test_attributes():
s = fake.session()
dataset = fake.dataset()
attrs = tc.dataset.attributes(s, dataset)
row_num = attrs[0]
assert row_num.name == "RowNum"
assert row_num.type == tc.attribute.type.STRING
geom = attrs[1]
assert geom.name == "geom"
assert isinstance(geom.type, tc.attribute.type.Record)
def test_materialize_async():
s = fake.session()
dataset = fake.dataset()
op = tc.dataset._materialize_async(s, dataset)
assert op.type == "SPARK"
assert op.description == "Materialize views to Elastic"
assert op.status == {
"state": "PENDING",
"startTime": "",
"endTime": "",
"message": "Job has not yet been submitted to Spark",
}
def test_upsert_index_as_primary_key():
s = fake.session()
dataset = fake.dataset()
df = pd.DataFrame(
_records_json_2,
index=[record["primary_key"] for record in _records_with_keys_json_2],
)
df.index.name = "primary_key"
response = tc.dataframe.upsert(s,
dataset,
df,
primary_key_name="primary_key")
assert response == _response_json
def test_from_resource_id():
s = fake.session()
dataset = fake.dataset()
attrs = tuple(
[
tc.SubAttribute(
name=str(i),
is_nullable=True,
type=tc.attribute.type.Array(tc.attribute.type.STRING),
)
for i in range(4)
]
)
attr = tc.attribute.from_resource_id(s, dataset, "attr")
assert attr.name == "attr"
assert not attr.is_nullable
assert isinstance(attr.type, tc.attribute.type.Record)
assert attr.type.attributes == attrs
def test_delete_infer_primary_key():
s = fake.session()
dataset = fake.dataset()
response = tc.record.delete(s, dataset, _records_json)
assert response == _response_json
def test_delete_dataset_not_found():
s = fake.session()
dataset = fake.dataset()
with pytest.raises(tc.dataset.NotFound):
tc.dataset.delete(s, dataset)
def test_create_reserved_attribute_name():
s = fake.session()
dataset = fake.dataset()
with pytest.raises(tc.attribute.ReservedName):
tc.attribute.create(s, dataset, name="clusterId", is_nullable=False)
def test_create_attribute_exists():
s = fake.session()
dataset = fake.dataset()
with pytest.raises(tc.attribute.AlreadyExists):
tc.attribute.create(s, dataset, name="attr", is_nullable=False)
def test_from_resource_id_attribute_not_found():
s = fake.session()
dataset = fake.dataset()
with pytest.raises(tc.attribute.NotFound):
tc.attribute.from_resource_id(s, dataset, "attr")
def test_stream():
s = fake.session()
dataset = fake.dataset()
records = tc.record.stream(s, dataset)
assert list(records) == _records_json
def test_delete_all():
s = fake.session()
dataset = fake.dataset()
tc.record.delete_all(s, dataset)
def test_delete_cascading():
s = fake.session()
dataset = fake.dataset()
tc.dataset.delete(s, dataset, cascade=True)
def test_delete():
s = fake.session()
dataset = fake.dataset()
tc.dataset.delete(s, dataset)
