Remove unused masked keyword in column_empty (#17530)

Follow up to #16715.

Now that the usages of the `masked` keyword in RAPIDS have been address (rapidsai/cuspatial#1496 is the only one I could find), I think we can remove this keyword all together in this method

Authors:
  - Matthew Roeschke (https://github.com/mroeschke)

Approvers:
  - Vyas Ramasubramani (https://github.com/vyasr)

URL: #17530

python/cudf/cudf/core/column/categorical.py

@@ -1193,7 +1193,7 @@ def _concat(
                 f"size > {libcudf.MAX_COLUMN_SIZE_STR}"
             )
         elif newsize == 0:
-            codes_col = column.column_empty(0, head.codes.dtype, masked=True)
+            codes_col = column.column_empty(0, head.codes.dtype)
         else:
             codes_col = column.concat_columns(codes)  # type: ignore[arg-type]
 

python/cudf/cudf/core/column/column.py

@@ -551,7 +551,7 @@ def slice(self, start: int, stop: int, stride: int | None = None) -> Self:
         if stop < 0 and not (stride < 0 and stop == -1):
             stop = stop + len(self)
         if (stride > 0 and start >= stop) or (stride < 0 and start <= stop):
-            return cast(Self, column_empty(0, self.dtype, masked=True))
+            return cast(Self, column_empty(0, self.dtype))
         # compute mask slice
         if stride == 1:
             return libcudf.copying.column_slice(self, [start, stop])[
@@ -1054,7 +1054,7 @@ def astype(self, dtype: Dtype, copy: bool = False) -> ColumnBase:
             if self.dtype == dtype:
                 result = self
             else:
-                result = column_empty(0, dtype=dtype, masked=self.nullable)
+                result = column_empty(0, dtype=dtype)
         elif dtype == "category":
             # TODO: Figure out why `cudf.dtype("category")`
             # astype's different than just the string
@@ -1625,7 +1625,6 @@ def _has_any_nan(arbitrary: pd.Series | np.ndarray) -> bool:
 def column_empty(
     row_count: int,
     dtype: Dtype = "object",
-    masked: bool = False,
     for_numba: bool = False,
 ) -> ColumnBase:
     """
@@ -1642,9 +1641,6 @@ def column_empty(
     dtype : Dtype
         Type of the column.
 
-    masked : bool
-        Unused.
-
     for_numba : bool, default False
         If True, don't allocate a mask as it's not supported by numba.
     """
@@ -2420,7 +2416,7 @@ def concat_columns(objs: "MutableSequence[ColumnBase]") -> ColumnBase:
     """Concatenate a sequence of columns."""
     if len(objs) == 0:
         dtype = cudf.dtype(None)
-        return column_empty(0, dtype=dtype, masked=True)
+        return column_empty(0, dtype=dtype)
 
     # If all columns are `NumericalColumn` with different dtypes,
     # we cast them to a common dtype.
@@ -2467,7 +2463,7 @@ def concat_columns(objs: "MutableSequence[ColumnBase]") -> ColumnBase:
             f"size > {libcudf.MAX_COLUMN_SIZE_STR}"
         )
     elif newsize == 0:
-        return column_empty(0, head.dtype, masked=True)
+        return column_empty(0, head.dtype)
 
     # Filter out inputs that have 0 length, then concatenate.
     objs_with_len = [o for o in objs if len(o)]

python/cudf/cudf/core/column/datetime.py

@@ -598,14 +598,12 @@ def strftime(self, format: str) -> cudf.core.column.StringColumn:
         if len(self) == 0:
             return cast(
                 cudf.core.column.StringColumn,
-                column.column_empty(0, dtype="object", masked=False),
+                column.column_empty(0, dtype="object"),
             )
         if format in _DATETIME_SPECIAL_FORMATS:
             names = as_column(_DATETIME_NAMES)
         else:
-            names = cudf.core.column.column_empty(
-                0, dtype="object", masked=False
-            )
+            names = column.column_empty(0, dtype="object")
         return string._datetime_to_str_typecast_functions[self.dtype](
             self, format, names
         )

python/cudf/cudf/core/column/numerical_base.py

@@ -139,7 +139,7 @@ def quantile(
             result = cast(
                 NumericalBaseColumn,
                 cudf.core.column.column_empty(
-                    row_count=len(q), dtype=self.dtype, masked=True
+                    row_count=len(q), dtype=self.dtype
                 ),
             )
         else:

python/cudf/cudf/core/column/string.py

@@ -5855,7 +5855,7 @@ def strptime(
                 f"dtype must be datetime or timedelta type, not {dtype}"
             )
         elif self.null_count == len(self):
-            return column.column_empty(len(self), dtype=dtype, masked=True)  # type: ignore[return-value]
+            return column.column_empty(len(self), dtype=dtype)  # type: ignore[return-value]
         elif (self == "None").any():
             raise ValueError(
                 "Cannot convert `None` value to datetime or timedelta."

python/cudf/cudf/core/column/timedelta.py

@@ -294,7 +294,7 @@ def strftime(self, format: str) -> cudf.core.column.StringColumn:
         if len(self) == 0:
             return cast(
                 cudf.core.column.StringColumn,
-                column.column_empty(0, dtype="object", masked=False),
+                column.column_empty(0, dtype="object"),
             )
         else:
             return string._timedelta_to_str_typecast_functions[self.dtype](

python/cudf/cudf/core/dataframe.py

@@ -774,9 +774,7 @@ def __init__(
                 label_dtype = getattr(columns, "dtype", None)
                 self._data = ColumnAccessor(
                     {
-                        k: column.column_empty(
-                            len(self), dtype="object", masked=True
-                        )
+                        k: column_empty(len(self), dtype="object")
                         for k in columns
                     },
                     level_names=tuple(columns.names)
@@ -979,8 +977,8 @@ def _init_from_series_list(self, data, columns, index):
         if columns is not None:
             for col_name in columns:
                 if col_name not in self._data:
-                    self._data[col_name] = column.column_empty(
-                        row_count=len(self), dtype=None, masked=True
+                    self._data[col_name] = column_empty(
+                        row_count=len(self), dtype=None
                     )
             self._data._level_names = (
                 tuple(columns.names)
@@ -1031,11 +1029,7 @@ def _init_from_list_like(self, data, index=None, columns=None):
             data = list(itertools.zip_longest(*data))
 
             if columns is not None and len(data) == 0:
-                data = [
-                    cudf.core.column.column_empty(row_count=0, dtype=None)
-                    for _ in columns
-                ]
-
+                data = [column_empty(row_count=0, dtype=None) for _ in columns]
             for col_name, col in enumerate(data):
                 self._data[col_name] = column.as_column(col)
             self._data.rangeindex = True
@@ -1074,9 +1068,8 @@ def _init_from_dict_like(
                 # the provided index, so we need to return a masked
                 # array of nulls if an index is given.
                 empty_column = functools.partial(
-                    cudf.core.column.column_empty,
-                    row_count=(0 if index is None else len(index)),
-                    masked=index is not None,
+                    column_empty,
+                    row_count=0 if index is None else len(index),
                 )
 
             data = {
@@ -1421,7 +1414,7 @@ def __setitem__(self, arg, value):
                         new_columns = (
                             value
                             if key == arg
-                            else column.column_empty(
+                            else column_empty(
                                 row_count=length, dtype=col.dtype
                             )
                             for key, col in self._column_labels_and_values
@@ -3373,7 +3366,7 @@ def _insert(self, loc, name, value, nan_as_null=None, ignore_index=True):
                 if num_cols != 0:
                     ca = self._data._from_columns_like_self(
                         (
-                            column.column_empty(row_count=length, dtype=dtype)
+                            column_empty(row_count=length, dtype=dtype)
                             for _, dtype in self._dtypes
                         ),
                         verify=False,
@@ -3479,7 +3472,7 @@ def diff(self, periods=1, axis=0):
         if abs(periods) > len(self):
             df = cudf.DataFrame._from_data(
                 {
-                    name: column_empty(len(self), dtype=dtype, masked=True)
+                    name: column_empty(len(self), dtype=dtype)
                     for name, dtype in zip(self._column_names, self.dtypes)
                 }
             )
@@ -3859,9 +3852,7 @@ def agg(self, aggs, axis=None):
                 result = DataFrame(index=idxs, columns=cols)
                 for key in aggs.keys():
                     col = self[key]
-                    col_empty = column_empty(
-                        len(idxs), dtype=col.dtype, masked=True
-                    )
+                    col_empty = column_empty(len(idxs), dtype=col.dtype)
                     ans = cudf.Series._from_column(
                         col_empty, index=cudf.Index(idxs)
                     )
@@ -6177,9 +6168,7 @@ def quantile(
                         quant_index=False,
                     )._column
                     if len(res) == 0:
-                        res = column.column_empty(
-                            row_count=len(qs), dtype=ser.dtype
-                        )
+                        res = column_empty(row_count=len(qs), dtype=ser.dtype)
                     result[k] = res
             result = DataFrame._from_data(result)
 
@@ -7333,9 +7322,7 @@ def unnamed_group_generator():
             )
 
             all_nulls = functools.cache(
-                functools.partial(
-                    column_empty, self.shape[0], common_type, masked=True
-                )
+                functools.partial(column_empty, self.shape[0], common_type)
             )
 
             # homogenize the dtypes of the columns
@@ -8582,7 +8569,7 @@ def _cast_cols_to_common_dtypes(col_idxs, list_of_columns, dtypes, categories):
             # If column not in this df, fill with an all-null column
             if idx >= len(cols) or cols[idx] is None:
                 n = len(next(x for x in cols if x is not None))
-                cols[idx] = column_empty(row_count=n, dtype=dtype, masked=True)
+                cols[idx] = column_empty(row_count=n, dtype=dtype)
             else:
                 # If column is categorical, rebase the codes with the
                 # combined categories, and cast the new codes to the

python/cudf/cudf/core/dtypes.py

@@ -189,9 +189,7 @@ def categories(self) -> cudf.Index:
         Index(['b', 'a'], dtype='object')
         """
         if self._categories is None:
-            col = cudf.core.column.column_empty(
-                0, dtype="object", masked=False
-            )
+            col = cudf.core.column.column_empty(0, dtype="object")
         else:
             col = self._categories
         return cudf.Index._from_column(col)

python/cudf/cudf/core/groupby/groupby.py

@@ -493,9 +493,7 @@ def size(self):
         """
         Return the size of each group.
         """
-        col = cudf.core.column.column_empty(
-            len(self.obj), "int8", masked=False
-        )
+        col = cudf.core.column.column_empty(len(self.obj), "int8")
         result = (
             cudf.Series._from_column(col, name=getattr(self.obj, "name", None))
             .groupby(self.grouping, sort=self._sort, dropna=self._dropna)
@@ -523,7 +521,8 @@ def cumcount(self, ascending: bool = True):
         return (
             cudf.Series._from_column(
                 cudf.core.column.column_empty(
-                    len(self.obj), "int8", masked=False
+                    len(self.obj),
+                    "int8",
                 ),
                 index=self.obj.index,
             )

python/cudf/cudf/core/index.py

@@ -336,7 +336,7 @@ def _values(self) -> ColumnBase:
         if len(self) > 0:
             return column.as_column(self._range, dtype=self.dtype)
         else:
-            return column.column_empty(0, masked=False, dtype=self.dtype)
+            return column.column_empty(0, dtype=self.dtype)
 
     def _clean_nulls_from_index(self) -> Self:
         return self

python/cudf/cudf/core/indexed_frame.py

@@ -3851,7 +3851,6 @@ def _reindex(
                 if name in df._data
                 else cudf.core.column.column.column_empty(
                     dtype=dtypes.get(name, np.float64),
-                    masked=True,
                     row_count=len(index),
                 )
             )

python/cudf/cudf/io/parquet.py

@@ -1139,7 +1139,6 @@ def _parquet_to_frame(
                     dfs[-1][name] = column_empty(
                         row_count=_len,
                         dtype=_dtype,
-                        masked=True,
                     )
                 else:
                     dfs[-1][name] = as_column(