def testDictList(self):
data = [{'Cluster': [
'_id',
'cluster.prop1.value',
'_label'],
'Organization': ['org.prop2.value',
'_id',
'_label'],
'Address': ['addr.prop1.value',
'addr.prop2.value',
'_label',
'_id'],
'Job': [
'_id',
'_label'],
'Person': ['cluster.prop1.value',
'cluster.prop2.value',
'cluster.prop3.value',
'_id',
'_label'],
'Source': [
'_id'],
'Nothing': []
}]
df = to_df(data)
self.assertEqual(df['Nothing'].values[0], None)
self.assertEqual(df['Source'].values[0], '_id')
self.assertListEqual(df['Job'].values[0], [
'_id',
'_label'])
def testNestedFields(self):
traversal_result = [{
"T.id": "T1",
"T.label": "node"
}, {
"T.id": "T2",
"T.label": "node",
"nested": {
"field1": [0]
}
}, {
"T.id": "T3",
"T.label": "node",
"nested": {
"field1": [1]
}
}, {
"T.id": "T4",
"T.label": "node"
}]
df = to_df(traversal_result)
test_df = pd.DataFrame(columns=['T.id', 'T.label', 'nested.field1'],
data=[['T1', 'node', np.nan],
['T2', 'node', 0.0], ['T3', 'node', 1.0],
['T4', 'node', np.nan]],
index=[0, 1, 2, 3])
self.assertTrue(
test_df.equals(df),
"Dictionary with nested field does not transform properly.")
def testProfiling(self):
data = [{
'@type': 'g:TraversalMetrics',
'@value': {
'dur':
0.8397500000000001,
'metrics': [{
'@type': 'g:Metrics',
'@value': {
'dur': 0.210785,
'counts': {
'traverserCount': 1,
'elementCount': 1
},
'name': 'NeptuneGraphQueryStep(Vertex)',
'annotations': {
'percentDur': 25.10092289371837
},
'id': '8.0.0()'
}
}, {
'@type': 'g:Metrics',
'@value': {
'dur': 0.628965,
'counts': {
'traverserCount': 1,
'elementCount': 1
},
'name': 'PropertyMapStep(value)',
'annotations': {
'percentDur': 74.89907710628164
},
'id': '4.0.0()'
}
}]
}
}]
df = to_df(data)
self.assertAlmostEqual(df.iloc[0, 0], 25.10092289371837, 5,
"DataFrame format is not correct.")
self.assertAlmostEqual(df.loc[1, "annotations.percentDur"], 74.89908,
4, "Profile data not correctly flattened.")
def testNoProfiling(self):
data = [{
'@type': 'g:TraversalMetrics',
'@value': {
'dur':
0.8397500000000001,
'metrics': [{
'@type': 'g:Metrics',
'@value': {
'dur': 0.210785,
'counts': {
'traverserCount': 1,
'elementCount': 1
},
'name': 'NeptuneGraphQueryStep(Vertex)',
'annotations': {
'percentDur': 25.10092289371837
},
'id': '8.0.0()'
}
}, {
'@type': 'g:Metrics',
'@value': {
'dur': 0.628965,
'counts': {
'traverserCount': 1,
'elementCount': 1
},
'name': 'PropertyMapStep(value)',
'annotations': {
'percentDur': 74.89907710628164
},
'id': '4.0.0()'
}
}]
}
}]
df = to_df(data, detect_profiling=False)
self.assertEqual(df.loc[0, '@type'], 'g:TraversalMetrics',
"DataFrame format is not correct.")
def testDeprecatedParameter(self):
traversal = wrap_content_as_traversal(Vertex("v1"), Vertex("v2"), Vertex("v3"))
with self.assertRaises(DeprecationWarning) as context:
to_df(traversal, keep_first_only=False)
def testColumns(self):
traversal = wrap_content_as_traversal(Vertex("v1"), Vertex("v2"), Vertex("v3"))
df = to_df(traversal)
self.assertListEqual(df.columns.values.tolist(), ['id', 'label'], "Incorrect field names extracted.")
def testToListCall(self):
traversal = wrap_content_as_traversal(Vertex("v1"), Vertex("v2"), Vertex("v3"))
df = to_df(traversal)
self.assertEqual(len(df.index), 3, "DataFrame is not populated properly.")
def testEmptyResult(self):
traversal_result = []
df = to_df(traversal_result)
self.assertTrue(pd.DataFrame().equals(df), "Empty result does not yield empty DataFrame.")
