jorisvandenbossche commented on code in PR #36538:
URL: https://github.com/apache/arrow/pull/36538#discussion_r1255670641
##########
python/pyarrow/tests/test_dataset.py:
##########
@@ -4551,20 +4553,32 @@ def test_write_dataset_parquet(tempdir):
file_paths = list(base_dir.rglob("*"))
expected_paths = [base_dir / "part-0.parquet"]
assert set(file_paths) == set(expected_paths)
- # check Table roundtrip
+ # check Table roundtrip with default version
result = ds.dataset(base_dir, format="parquet").to_table()
assert result.equals(table)
# using custom options
for version in ["1.0", "2.4", "2.6"]:
format = ds.ParquetFileFormat()
opts = format.make_write_options(version=version)
+ assert "<pyarrow.dataset.ParquetFileWriteOptions" in repr(opts)
base_dir = tempdir / 'parquet_dataset_version{0}'.format(version)
ds.write_dataset(table, base_dir, format=format, file_options=opts)
meta = pq.read_metadata(base_dir / "part-0.parquet")
expected_version = "1.0" if version == "1.0" else "2.6"
assert meta.format_version == expected_version
+ # ensure version is actually honored based on supported datatypes
+ result = ds.dataset(base_dir, format="parquet").to_table()
+ schema = table.schema
+ if version == "1.0":
+ # uint32 is written as int64
+ schema = schema.set(0, schema.field(0).with_type(pa.int64()))
+ if version in ("1.0", "2.4"):
+ schema = schema.set(1,
schema.field(1).with_type(pa.timestamp("us")))
Review Comment:
I created the dataset with nanoseconds, so `schema` has nanoseconds for 2.6
(I only change it to microseconds for the older versions), so it is also being
tested here
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
