def test_flatten_with_array(self):
sr = SidecarMap()
sidecar1 = {"a": {"c": ["blech3", "blech3a"], "d": "blech4", "e": "blech5"}, "b": "blech2"}
try:
sr.flatten(sidecar1)
except HedFileError:
pass
except Exception:
self.fail('process threw the wrong exception when array included in JSON')
else:
self.fail('process should have thrown a HedFileError when array included in JSON')
def test_flatten_sidecar_with_empty(self):
sr = SidecarMap()
# top level value is empty
sidecar1 = {"a": {"c": "blech3", "d": "blech4", "e": "blech5"}, "b": "",
"af": {"c": {"af": "blech6"}, "df_new": "blech7"}}
df1 = sr.flatten(sidecar1)
self.assertEqual(12, len(df1), "flatten should return a dataframe with correct number of rows")
self.assertEqual(2, len(df1.columns), "flatten should return a dataframe with 2 columns")
# other value is empty
sidecar2 = {"a": {"c": "blech3", "d": "blech4", "e": "blech5"}, "b": "blech2",
"af": {"c": {"af": "blech6"}, "df_new": ""}}
df2 = sr.flatten(sidecar2)
self.assertEqual(12, len(df2), "flatten should return a dataframe with correct number of rows")
self.assertEqual(2, len(df2.columns), "flatten should return a dataframe with 2 columns")
def sidecar_flatten(json_sidecar):
"""Converts a sidecar from long to short unless unless the command is not COMMAND_TO_LONG then converts to short
Parameters
----------
json_sidecar: Sidecar
Previously created Sidecar
Returns
-------
dict
A downloadable dictionary file or a file containing warnings
"""
json_string = json_sidecar.get_as_json_string()
sidecar = json.loads(json_string)
sr = SidecarMap()
df = sr.flatten(sidecar)
data = df.to_csv(None, sep='\t', index=False, header=True)
display_name = json_sidecar.name
file_name = generate_filename(display_name,
name_suffix='flattened',
extension='.tsv')
return {
base_constants.COMMAND: base_constants.COMMAND_FLATTEN,
'data': data,
'output_display_name': file_name,
'msg_category': 'success',
'msg': f'JSON sidecar {display_name} was successfully flattened'
}
def test_unflatten(self):
sr = SidecarMap()
sidecar1 = {"a": "blech1", "d": "blech4", "e": "blech5"}
df1 = sr.flatten(sidecar1)
unflat1 = sr.unflatten(df1)
self.assertEqual(sidecar1, unflat1, "unflatten should unflatten when sidecar unnested")
sidecar2 = {"a": {"c1": "blech3", "c2": "blech3a"}, "d": "apple"}
df2 = sr.flatten(sidecar2)
unflat2 = sr.unflatten(df2)
self.assertEqual(sidecar2, unflat2, "unflatten should unflatten when sidecar has single dictionary")
sidecar3 = {"b": "banana", "a": {"c1": "blech3", "c2": "blech3a"}, "d": "apple"}
df3 = sr.flatten(sidecar3)
unflat3 = sr.unflatten(df3)
self.assertEqual(sidecar3, unflat3, "unflatten should unflatten when sidecar has embedded dictionary")
sidecar4 = {"a": {"c": {"c1": "blech3", "c2": "blech3a"}, "d": "blech4", "e": "blech5"}, "b": "blech2"}
df4 = sr.flatten(sidecar4)
unflat4 = sr.unflatten(df4)
self.assertEqual(sidecar4, unflat4, "unflatten should unflatten when sidecar has nested dictionaries")
def test_flatten(self):
sr = SidecarMap()
sidecar1 = {"a": {"c": {"c1": "blech3", "c2": "blech3a"}, "d": "blech4", "e": "blech5"}, "b": "blech2"}
df1 = sr.flatten(sidecar1)
self.assertEqual(9, len(df1), "flatten should return a dataframe with correct number of rows")
self.assertEqual(2, len(df1.columns), "flatten should return a dataframe with 2 columns")