def test_is_match(self):
"""
Determine if the parquet file can be automatically identified from
byte stream or file path
"""
for input_file in self.file_or_buf_list:
self.assertTrue(ParquetData.is_match(input_file["path"]))
def test_is_structured(self):
# Default construction
data = ParquetData()
self.assertTrue(data.is_structured)
# With option specifying dataframe as data_format
data = ParquetData(options={"data_format": "dataframe"})
self.assertTrue(data.is_structured)
# With option specifying records as data_format
data = ParquetData(options={"data_format": "records"})
self.assertFalse(data.is_structured)
# With option specifying json as data_format
data = ParquetData(options={"data_format": "json"})
self.assertFalse(data.is_structured)
def test_mixed_non_string_col(self):
"""
Determine if parquet can handle mixed non-string column types.
"""
test_file = os.path.join(test_root_path, "data", "parquet",
"mixed_datetime_data_col.parquet")
parq_data = ParquetData(test_file)
# assert str and not bytes
self.assertIsInstance(parq_data.data["col2"][1], str)
self.assertIsInstance(parq_data.data["col2"][3], str)
# assert no 'b"data"' encapsulated, just 'data'
self.assertNotIn('b"', parq_data.data["col2"][1])
self.assertNotIn("b'", parq_data.data["col2"][1])
self.assertNotIn('b"', parq_data.data["col2"][3])
self.assertNotIn("b'", parq_data.data["col2"][3])
def test_file_encoding(self):
"""Tests to ensure file_encoding set to None"""
for input_file in self.file_or_buf_list:
data = ParquetData(input_file["path"])
self.assertIsNone(data.file_encoding)