AlenkaF commented on code in PR #50102:
URL: https://github.com/apache/arrow/pull/50102#discussion_r3372709732
##########
python/pyarrow/tests/test_compute.py:
##########
@@ -1297,6 +1297,70 @@ def test_replace_with_mask_null_type():
assert result.to_pylist() == [None]
[email protected]("arr,mask,replacements,expected", [
Review Comment:
```suggestion
@pytest.mark.parametrize("arr, mask, replacements, expected", [
```
Same for others.
##########
python/pyarrow/tests/test_compute.py:
##########
@@ -1297,6 +1297,70 @@ def test_replace_with_mask_null_type():
assert result.to_pylist() == [None]
[email protected]("arr,mask,replacements,expected", [
+ # Basic replacement with array mask
+ (pa.array([1, 2, 3, 4, 5]), pa.array([True, False, True, False, True]),
+ pa.array([10, 20, 30]), pa.array([10, 2, 20, 4, 30])),
+ # Scalar mask True/False
+ (pa.array([1, 2, 3]), True, pa.array([10, 20, 30]), pa.array([10, 20,
30])),
+ (pa.array([1, 2, 3]), False, pa.array([], type=pa.int64()), pa.array([1,
2, 3])),
+ # Scalar replacement
+ (pa.array([1, 2, 3, 4]), pa.array([True, False, True, False]),
+ pa.scalar(99), pa.array([99, 2, 99, 4])),
+ # Null handling in input array
+ (pa.array([1, None, 3, None, 5]), pa.array([False, True, False, True,
True]),
+ pa.array([10, 20, 30]), pa.array([1, 10, 3, 20, 30])),
+ # Null handling in mask
+ (pa.array([1, 2, 3, 4, 5, 6]), pa.array([False, False, None, None, True,
True]),
+ pa.array([10, None]), pa.array([1, 2, None, None, 10, None])),
+ # String type
+ (pa.array(['a', 'b', 'c', 'd']), pa.array([True, False, True, False]),
+ pa.array(['x', 'y']), pa.array(['x', 'b', 'y', 'd'])),
+ # Float type
+ (pa.array([1.1, 2.2, 3.3, 4.4]), pa.array([True, False, True, False]),
+ pa.array([10.5, 20.5]), pa.array([10.5, 2.2, 20.5, 4.4])),
+])
+def test_replace_with_mask(arr, mask, replacements, expected):
+ """Test replace_with_mask with various data types."""
+ result = pc.replace_with_mask(arr, mask, replacements)
+ assert result.equals(expected)
+
+
[email protected]("arr,mask,replacements,expected", [
+ # ChunkedArray with multiple chunks
+ (pa.chunked_array([[1, 2, 3], [4, 5, 6]]),
+ pa.array([True, False, False, False, True, False]),
+ pa.array([10, 20]),
+ pa.chunked_array([[10, 2, 3], [4, 20, 6]])),
+ # ChunkedArray with empty chunks
+ (pa.chunked_array([[1, 2], [], [3, 4]]),
+ pa.array([True, False, True, False]),
+ pa.array([10, 20]),
+ pa.chunked_array([[10, 2], [20, 4]])),
+])
+def test_replace_with_mask_chunked_array(arr, mask, replacements, expected):
+ """Test replace_with_mask with ChunkedArray inputs."""
+ result = pc.replace_with_mask(arr, mask, replacements)
+ assert result.equals(expected)
+
+
[email protected]("arr,mask,replacements,error_match", [
+ # Replacement count does not match the number of true values in the mask
+ (pa.array([1, 2, 3]), pa.array([True, True, False]),
+ pa.array([10]), "expected 2.*but got 1"),
+ # Mask length does not match input array length
+ (pa.array([1, 2, 3]), pa.array([True, False]), pa.array([10]), None),
+])
+def test_replace_with_mask_errors(arr, mask, replacements, error_match):
Review Comment:
I would remove the if branch and the whole parametrization here and do a
clean test where array construction and `pytest.raises` check comes one after
the other.
--
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]
