This is an automated email from the ASF dual-hosted git repository.
bkietz 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 3360989 ARROW-8315: [Python] Fix dataset tests on Python 3.5
3360989 is described below
commit 3360989e04e99a4fcb50d7e068b6de469ca30716
Author: Antoine Pitrou <[email protected]>
AuthorDate: Thu Apr 2 13:46:37 2020 -0400
ARROW-8315: [Python] Fix dataset tests on Python 3.5
Do not rely on well-defined dict ordering.
Closes #6814 from pitrou/ARROW-8315-dataset-tests-py35
Authored-by: Antoine Pitrou <[email protected]>
Signed-off-by: Benjamin Kietzman <[email protected]>
---
python/pyarrow/tests/test_dataset.py | 10 ++++++----
python/pyarrow/tests/test_schema.py | 6 +-----
2 files changed, 7 insertions(+), 9 deletions(-)
diff --git a/python/pyarrow/tests/test_dataset.py
b/python/pyarrow/tests/test_dataset.py
index 037ba52..dc28512 100644
--- a/python/pyarrow/tests/test_dataset.py
+++ b/python/pyarrow/tests/test_dataset.py
@@ -625,7 +625,8 @@ def _create_dataset_for_fragments(tempdir, chunk_size=None):
import pyarrow.parquet as pq
table = pa.table(
- {'f1': range(8), 'f2': [1] * 8, 'part': ['a'] * 4 + ['b'] * 4}
+ [range(8), [1] * 8, ['a'] * 4 + ['b'] * 4],
+ names=['f1', 'f2', 'part']
)
# write_to_dataset currently requires pandas
pq.write_to_dataset(table, str(tempdir / "test_parquet_dataset"),
@@ -1176,13 +1177,14 @@ def test_specified_schema(tempdir):
# Specifying schema with missing column
schema = pa.schema([('a', 'int64')])
- expected = pa.table({'a': [1, 2, 3]})
+ expected = pa.table([[1, 2, 3]], names=['a'])
_check_dataset(schema, expected)
# Specifying schema with additional column
schema = pa.schema([('a', 'int64'), ('c', 'int32')])
- expected = pa.table({'a': [1, 2, 3],
- 'c': pa.array([None, None, None], type='int32')})
+ expected = pa.table([[1, 2, 3],
+ pa.array([None, None, None], type='int32')],
+ names=['a', 'c'])
_check_dataset(schema, expected)
# Specifying with incompatible schema
diff --git a/python/pyarrow/tests/test_schema.py
b/python/pyarrow/tests/test_schema.py
index d2b1bd2..d4cbfb3 100644
--- a/python/pyarrow/tests/test_schema.py
+++ b/python/pyarrow/tests/test_schema.py
@@ -261,8 +261,7 @@ sapien. Quisque pretium vestibulum urna eu vehicula."""
metadata={"key1": "value1"}),
pa.field("bar", "string", True,
metadata={"key3": "value3"})],
- metadata={"key2": "value2",
- "lorem": lorem})
+ metadata={"lorem": lorem})
assert my_schema.to_string() == """\
foo: int32 not null
@@ -272,7 +271,6 @@ bar: string
-- field metadata --
key3: 'value3'
-- schema metadata --
-key2: 'value2'
lorem: '""" + lorem[:65] + "' + " + str(len(lorem) - 65)
# Metadata that exactly fits
@@ -292,7 +290,6 @@ bar: string
-- field metadata --
key3: 'value3'
-- schema metadata --
-key2: 'value2'
lorem: '{}'""".format(lorem)
assert my_schema.to_string(truncate_metadata=False,
@@ -300,7 +297,6 @@ lorem: '{}'""".format(lorem)
foo: int32 not null
bar: string
-- schema metadata --
-key2: 'value2'
lorem: '{}'""".format(lorem)
assert my_schema.to_string(truncate_metadata=False,
