MCPcopy
hub / github.com/huggingface/datasets / data_dir_with_arrow

Function data_dir_with_arrow

tests/test_load.py:94–109  ·  view source on GitHub ↗
(tmp_path)

Source from the content-addressed store, hash-verified

92
93@pytest.fixture
94def data_dir_with_arrow(tmp_path):
95 data_dir = tmp_path / "data_dir"
96 data_dir.mkdir()
97 output_train = os.path.join(data_dir, "train.arrow")
98 with ArrowWriter(path=output_train) as writer:
99 writer.write_table(pa.Table.from_pydict({"col_1": ["foo"] * 10}))
100 num_examples, num_bytes = writer.finalize()
101 assert num_examples == 10
102 assert num_bytes > 0
103 output_test = os.path.join(data_dir, "test.arrow")
104 with ArrowWriter(path=output_test) as writer:
105 writer.write_table(pa.Table.from_pydict({"col_1": ["bar"] * 10}))
106 num_examples, num_bytes = writer.finalize()
107 assert num_examples == 10
108 assert num_bytes > 0
109 return str(data_dir)
110
111
112@pytest.fixture

Callers

nothing calls this directly

Calls 5

ArrowWriterClass · 0.90
mkdirMethod · 0.80
write_tableMethod · 0.80
from_pydictMethod · 0.80
finalizeMethod · 0.80

Tested by

no test coverage detected