Skip to content

CoW: Avoid warnings in stata code #56392

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

Merged
merged 3 commits into from
Dec 15, 2023
Merged

CoW: Avoid warnings in stata code #56392

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
3c2e34b
CoW: Avoid warnings in stata code
phofl Dec 7, 2023
22fc104
CoW: Avoid warnings in stata code
phofl Dec 7, 2023
a9fcd91
Update
phofl Dec 8, 2023
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
22 changes: 9 additions & 13 deletions pandas/io/stata.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -345,10 +345,7 @@ def convert_delta_safe(base, deltas, unit) -> Series:
has_bad_values = False
if bad_locs.any():
has_bad_values = True
# reset cache to avoid SettingWithCopy checks (we own the DataFrame and the
# `dates` Series is used to overwrite itself in the DataFramae)
dates._reset_cacher()
dates[bad_locs] = 1.0 # Replace with NaT
dates._values[bad_locs] = 1.0 # Replace with NaT
dates = dates.astype(np.int64)

if fmt.startswith(("%tc", "tc")): # Delta ms relative to base
Expand Down Expand Up @@ -467,9 +464,9 @@ def g(x: datetime) -> int:
if bad_loc.any():
dates = Series(dates)
Copy link
Member

Choose a reason for hiding this comment

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

Suggested change
dates = Series(dates)

And then below could use the existing dates[bad_loc] ? (if we are otherwise just setting on the _values, there is not much point in converting to a Series?)

Copy link
Member

Choose a reason for hiding this comment

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

Ah, but from seeing the surrounding code, I assume dates is already a Series, so this wouldn't change anything (but can also be left out then, I assume)

Copy link
Member Author

Choose a reason for hiding this comment

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

Yes, we can probably remove it but we already have references, so it won't help with the warning

if lib.is_np_dtype(dates.dtype, "M"):
dates[bad_loc] = to_datetime(stata_epoch)
dates._values[bad_loc] = to_datetime(stata_epoch)
else:
dates[bad_loc] = stata_epoch
dates._values[bad_loc] = stata_epoch

if fmt in ["%tc", "tc"]:
d = parse_dates_safe(dates, delta=True)
Expand Down Expand Up @@ -599,9 +596,8 @@ def _cast_to_stata_types(data: DataFrame) -> DataFrame:
for col in data:
# Cast from unsupported types to supported types
is_nullable_int = isinstance(data[col].dtype, (IntegerDtype, BooleanDtype))
orig = data[col]
# We need to find orig_missing before altering data below
orig_missing = orig.isna()
orig_missing = data[col].isna()
if is_nullable_int:
missing_loc = data[col].isna()
if missing_loc.any():
Expand Down Expand Up @@ -1783,15 +1779,15 @@ def read(
for idx in valid_dtypes:
dtype = data.iloc[:, idx].dtype
if dtype not in (object_type, self._dtyplist[idx]):
data.iloc[:, idx] = data.iloc[:, idx].astype(dtype)
data.isetitem(idx, data.iloc[:, idx].astype(dtype))

data = self._do_convert_missing(data, convert_missing)

if convert_dates:
for i, fmt in enumerate(self._fmtlist):
if any(fmt.startswith(date_fmt) for date_fmt in _date_formats):
data.iloc[:, i] = _stata_elapsed_date_to_datetime_vec(
data.iloc[:, i], fmt
data.isetitem(
i, _stata_elapsed_date_to_datetime_vec(data.iloc[:, i], fmt)
)

if convert_categoricals and self._format_version > 108:
Expand Down Expand Up @@ -1866,7 +1862,7 @@ def _do_convert_missing(self, data: DataFrame, convert_missing: bool) -> DataFra
replacements[i] = replacement
if replacements:
for idx, value in replacements.items():
data.iloc[:, idx] = value
data.isetitem(idx, value)
return data

def _insert_strls(self, data: DataFrame) -> DataFrame:
Expand All @@ -1876,7 +1872,7 @@ def _insert_strls(self, data: DataFrame) -> DataFrame:
if typ != "Q":
continue
# Wrap v_o in a string to allow uint64 values as keys on 32bit OS
data.iloc[:, i] = [self.GSO[str(k)] for k in data.iloc[:, i]]
data.isetitem(i, [self.GSO[str(k)] for k in data.iloc[:, i]])
return data

def _do_select_columns(self, data: DataFrame, columns: Sequence[str]) -> DataFrame:
Expand Down
2 changes: 0 additions & 2 deletions pandas/tests/io/test_common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -289,8 +289,6 @@ def test_read_expands_user_home_dir(
):
reader(path)

# TODO(CoW-warn) avoid warnings in the stata reader code
@pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
@pytest.mark.parametrize(
"reader, module, path",
[
Expand Down
16 changes: 4 additions & 12 deletions pandas/tests/io/test_stata.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,11 +32,6 @@
read_stata,
)

# TODO(CoW-warn) avoid warnings in the stata reader code
pytestmark = pytest.mark.filterwarnings(
"ignore:Setting a value on a view:FutureWarning"
)


@pytest.fixture
def mixed_frame():
Expand Down Expand Up @@ -140,8 +135,7 @@ def test_read_dta1(self, file, datapath):

tm.assert_frame_equal(parsed, expected)

@pytest.mark.filterwarnings("always")
def test_read_dta2(self, datapath):
def test_read_dta2(self, datapath, warn_copy_on_write):
expected = DataFrame.from_records(
[
(
Expand Down Expand Up @@ -183,13 +177,11 @@ def test_read_dta2(self, datapath):
path2 = datapath("io", "data", "stata", "stata2_115.dta")
path3 = datapath("io", "data", "stata", "stata2_117.dta")

# TODO(CoW-warn) avoid warnings in the stata reader code
# once fixed -> remove `raise_on_extra_warnings=False` again
with tm.assert_produces_warning(UserWarning, raise_on_extra_warnings=False):
with tm.assert_produces_warning(UserWarning):
parsed_114 = self.read_dta(path1)
with tm.assert_produces_warning(UserWarning, raise_on_extra_warnings=False):
with tm.assert_produces_warning(UserWarning):
parsed_115 = self.read_dta(path2)
with tm.assert_produces_warning(UserWarning, raise_on_extra_warnings=False):
with tm.assert_produces_warning(UserWarning):
parsed_117 = self.read_dta(path3)
# FIXME: don't leave commented-out
# 113 is buggy due to limits of date format support in Stata
Expand Down