This is an automated email from the ASF dual-hosted git repository.
jonkeane pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow.git
The following commit(s) were added to refs/heads/master by this push:
new 2ffc10a43b ARROW-14632: [Python] Make write_dataset arguments
keyword-only
2ffc10a43b is described below
commit 2ffc10a43b2b9a397bfeba993993172082f9722b
Author: Austin Dickey <[email protected]>
AuthorDate: Wed Jun 1 17:22:17 2022 -0500
ARROW-14632: [Python] Make write_dataset arguments keyword-only
As a best practice, most of the optional configuration arguments in
`write_dataset()` should be keyword-only. This PR enforces that.
Closes #13289 from austin3dickey/ARROW-14632
Authored-by: Austin Dickey <[email protected]>
Signed-off-by: Jonathan Keane <[email protected]>
---
python/pyarrow/dataset.py | 2 +-
python/pyarrow/tests/test_dataset.py | 6 ++++++
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/python/pyarrow/dataset.py b/python/pyarrow/dataset.py
index 6c1b8db5a6..8ef3e2f7aa 100644
--- a/python/pyarrow/dataset.py
+++ b/python/pyarrow/dataset.py
@@ -801,7 +801,7 @@ def _ensure_write_partitioning(part, schema, flavor):
return part
-def write_dataset(data, base_dir, basename_template=None, format=None,
+def write_dataset(data, base_dir, *, basename_template=None, format=None,
partitioning=None, partitioning_flavor=None, schema=None,
filesystem=None, file_options=None, use_threads=True,
max_partitions=None, max_open_files=None,
diff --git a/python/pyarrow/tests/test_dataset.py
b/python/pyarrow/tests/test_dataset.py
index 0be01d2336..d2210c4b6c 100644
--- a/python/pyarrow/tests/test_dataset.py
+++ b/python/pyarrow/tests/test_dataset.py
@@ -1796,6 +1796,12 @@ def
test_dictionary_partitioning_outer_nulls_raises(tempdir):
ds.write_dataset(table, tempdir, format='ipc', partitioning=part)
+def test_positional_keywords_raises(tempdir):
+ table = pa.table({'a': ['x', 'y', None], 'b': ['x', 'y', 'z']})
+ with pytest.raises(TypeError):
+ ds.write_dataset(table, tempdir, "basename-{i}.arrow")
+
+
@pytest.mark.parquet
@pytest.mark.pandas
def test_read_partition_keys_only(tempdir):
