Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

fix: dataframe fillna with scalar. #1132

Merged
merged 5 commits into from
Nov 13, 2024
Merged

fix: dataframe fillna with scalar. #1132

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
81f498e
fix: dataframe fillna with string scalar.
Genesis929 Nov 5, 2024
85c4254
update type supports
Genesis929 Nov 6, 2024
557aa66
remove case that pandas has issue
Genesis929 Nov 6, 2024
fa14f28
Merge branch 'main' into fillna_fix_huanc
Genesis929 Nov 6, 2024
2e27dff
update annotation
Genesis929 Nov 12, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 5 additions & 2 deletions bigframes/dataframe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -734,7 +734,7 @@ def _apply_binop(
how: str = "outer",
reverse: bool = False,
):
if isinstance(other, (float, int, bool)):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I feel like just adding another type here is a narrow fix - what we really want to do is determine - is this type interpretable as a supported scalar. This could include datatime objects, decimal, etc as well. We should probably have a single definition of this somewhere

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added a LOCAL_SCALAR_TYPES constant, all types are from infer_literal_type function, should be supported.

if isinstance(other, bigframes.dtypes.LOCAL_SCALAR_TYPES):
return self._apply_scalar_binop(other, op, reverse=reverse)
elif isinstance(other, DataFrame):
return self._apply_dataframe_binop(other, op, how=how, reverse=reverse)
Expand All @@ -752,7 +752,10 @@ def _apply_binop(
)

def _apply_scalar_binop(
self, other: float | int, op: ops.BinaryOp, reverse: bool = False
self,
other: bigframes.dtypes.LOCAL_SCALAR_TYPE,
op: ops.BinaryOp,
reverse: bool = False,
) -> DataFrame:
if reverse:
expr = op.as_expr(
Expand Down
19 changes: 19 additions & 0 deletions bigframes/dtypes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,6 +59,25 @@
# Used when storing Null expressions
DEFAULT_DTYPE = FLOAT_DTYPE

LOCAL_SCALAR_TYPE = Union[
bool,
np.bool_,
int,
np.integer,
float,
np.floating,
decimal.Decimal,
str,
np.str_,
bytes,
np.bytes_,
datetime.datetime,
pd.Timestamp,
datetime.date,
datetime.time,
]
LOCAL_SCALAR_TYPES = typing.get_args(LOCAL_SCALAR_TYPE)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have a feeling we will just give up at a certain point as we add more types



# Will have a few dtype variants: simple(eg. int, string, bool), complex (eg. list, struct), and virtual (eg. micro intervals, categorical)
@dataclass(frozen=True)
Expand Down
17 changes: 12 additions & 5 deletions tests/system/small/test_dataframe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1020,13 +1020,20 @@ def test_df_interpolate(scalars_dfs):
)


def test_df_fillna(scalars_dfs):
@pytest.mark.parametrize(
shobsi marked this conversation as resolved.
Show resolved Hide resolved
"col, fill_value",
[
(["int64_col", "float64_col"], 3),
(["string_col"], "A"),
(["datetime_col"], pd.Timestamp("2023-01-01")),
],
)
def test_df_fillna(scalars_dfs, col, fill_value):
scalars_df, scalars_pandas_df = scalars_dfs
df = scalars_df[["int64_col", "float64_col"]].fillna(3)
bf_result = df.to_pandas()
pd_result = scalars_pandas_df[["int64_col", "float64_col"]].fillna(3)
bf_result = scalars_df[col].fillna(fill_value).to_pandas()
pd_result = scalars_pandas_df[col].fillna(fill_value)

pandas.testing.assert_frame_equal(bf_result, pd_result)
pd.testing.assert_frame_equal(bf_result, pd_result, check_dtype=False)


def test_df_replace_scalar_scalar(scalars_dfs):
Expand Down