MCPcopy Create free account
hub / github.com/apache/arrow / test_chunked_binary_error_message

Function test_chunked_binary_error_message

python/pyarrow/tests/test_feather.py:729–748  ·  view source on GitHub ↗
()

Source from the content-addressed store, hash-verified

727@pytest.mark.large_memory
728@pytest.mark.pandas
729def test_chunked_binary_error_message():
730 # ARROW-3058: As Feather does not yet support chunked columns, we at least
731 # make sure it's clear to the user what is going on
732
733 # 2^31 + 1 bytes
734 values = [b'x'] + [
735 b'x' * (1 << 20)
736 ] * 2 * (1 << 10)
737 df = pd.DataFrame({'byte_col': values})
738
739 # Works fine with version 2
740 buf = io.BytesIO()
741 write_feather(df, buf, version=2)
742 result = read_feather(pa.BufferReader(buf.getvalue()))
743 assert_frame_equal(result, df)
744
745 with pytest.raises(ValueError, match="'byte_col' exceeds 2GB maximum "
746 "capacity of a Feather binary column. This restriction "
747 "may be lifted in the future"):
748 write_feather(df, io.BytesIO(), version=1)
749
750
751def test_feather_without_pandas(tempdir, version):

Callers

nothing calls this directly

Calls 3

write_featherFunction · 0.90
read_featherFunction · 0.90
BufferReaderMethod · 0.80

Tested by

no test coverage detected