REF: rename _data-->_mgr

pandas/_libs/properties.pyx

@@ -57,10 +57,10 @@ cdef class AxisProperty:
             list axes
 
         if obj is None:
-            # Only instances have _data, not classes
+            # Only instances have _mgr, not classes
             return self
         else:
-            axes = obj._data.axes
+            axes = obj._mgr.axes
         return axes[self.axis]
 
     def __set__(self, obj, value):

pandas/_libs/reduction.pyx

@@ -148,7 +148,7 @@ cdef class Reducer:
                         object.__setattr__(cached_typ, 'index', self.index)
 
                     object.__setattr__(
-                        cached_typ._data._block, 'values', chunk)
+                        cached_typ._mgr._block, 'values', chunk)
                     object.__setattr__(cached_typ, 'name', name)
                     res = self.f(cached_typ)
                 else:
@@ -279,7 +279,7 @@ cdef class SeriesBinGrouper:
                     object.__setattr__(cached_ityp, '_index_data', islider.buf)
                     cached_ityp._engine.clear_mapping()
                     object.__setattr__(
-                        cached_typ._data._block, 'values', vslider.buf)
+                        cached_typ._mgr._block, 'values', vslider.buf)
                     object.__setattr__(cached_typ, '_index', cached_ityp)
                     object.__setattr__(cached_typ, 'name', name)
 
@@ -405,7 +405,7 @@ cdef class SeriesGrouper:
                         object.__setattr__(cached_ityp, '_data', islider.buf)
                         cached_ityp._engine.clear_mapping()
                         object.__setattr__(
-                            cached_typ._data._block, 'values', vslider.buf)
+                            cached_typ._mgr._block, 'values', vslider.buf)
                         object.__setattr__(cached_typ, '_index', cached_ityp)
                         object.__setattr__(cached_typ, 'name', name)
 
@@ -577,7 +577,7 @@ cdef class BlockSlider:
         self.dummy = frame[:0]
         self.index = self.dummy.index
 
-        self.blocks = [b.values for b in self.dummy._data.blocks]
+        self.blocks = [b.values for b in self.dummy._mgr.blocks]
 
         for x in self.blocks:
             util.set_array_not_contiguous(x)

pandas/_libs/src/ujson/python/objToJSON.c

@@ -326,7 +326,7 @@ static PyObject *get_sub_attr(PyObject *obj, char *attr, char *subAttr) {
 }
 
 static int is_simple_frame(PyObject *obj) {
-    PyObject *check = get_sub_attr(obj, "_data", "is_mixed_type");
+    PyObject *check = get_sub_attr(obj, "_mgr", "is_mixed_type");
     int ret = (check == Py_False);
 
     if (!check) {
@@ -984,7 +984,7 @@ void PdBlock_iterBegin(JSOBJ _obj, JSONTypeContext *tc) {
         goto BLKRET;
     }
 
-    blocks = get_sub_attr(obj, "_data", "blocks");
+    blocks = get_sub_attr(obj, "_mgr", "blocks");
     if (!blocks) {
         GET_TC(tc)->iterNext = NpyArr_iterNextNone;
         goto BLKRET;

pandas/core/algorithms.py

@@ -1710,7 +1710,7 @@ def take_nd(
         if arr.flags.f_contiguous and axis == arr.ndim - 1:
             # minor tweak that can make an order-of-magnitude difference
             # for dataframes initialized directly from 2-d ndarrays
-            # (s.t. df.values is c-contiguous and df._data.blocks[0] is its
+            # (s.t. df.values is c-contiguous and df._mgr.blocks[0] is its
             # f-contiguous transpose)
             out = np.empty(out_shape, dtype=dtype, order="F")
         else:

pandas/core/apply.py

@@ -166,7 +166,7 @@ def get_result(self):
         # ufunc
         elif isinstance(self.f, np.ufunc):
             with np.errstate(all="ignore"):
-                results = self.obj._data.apply("apply", func=self.f)
+                results = self.obj._mgr.apply("apply", func=self.f)
             return self.obj._constructor(
                 data=results, index=self.index, columns=self.columns, copy=False
             )

pandas/core/dtypes/generic.py

@@ -80,7 +80,7 @@ def _check(cls, inst):
 
 class _ABCGeneric(type):
     def __instancecheck__(cls, inst):
-        return hasattr(inst, "_data")
+        return hasattr(inst, "_mgr")
 
 
 ABCGeneric = _ABCGeneric("ABCGeneric", tuple(), {})

pandas/core/dtypes/missing.py

@@ -144,7 +144,7 @@ def _isna_new(obj):
     ):
         return _isna_ndarraylike(obj)
     elif isinstance(obj, ABCGeneric):
-        return obj._constructor(obj._data.isna(func=isna))
+        return obj._constructor(obj._mgr.isna(func=isna))
     elif isinstance(obj, list):
         return _isna_ndarraylike(np.asarray(obj, dtype=object))
     elif hasattr(obj, "__array__"):
@@ -172,7 +172,7 @@ def _isna_old(obj):
     elif isinstance(obj, (ABCSeries, np.ndarray, ABCIndexClass)):
         return _isna_ndarraylike_old(obj)
     elif isinstance(obj, ABCGeneric):
-        return obj._constructor(obj._data.isna(func=_isna_old))
+        return obj._constructor(obj._mgr.isna(func=_isna_old))
     elif isinstance(obj, list):
         return _isna_ndarraylike_old(np.asarray(obj, dtype=object))
     elif hasattr(obj, "__array__"):

pandas/core/frame.py

@@ -395,7 +395,7 @@ def __init__(self, data=None, index=None, columns=None, dtype=None, copy=False):
             dtype = self._validate_dtype(dtype)
 
         if isinstance(data, DataFrame):
-            data = data._data
+            data = data._mgr
 
         if isinstance(data, BlockManager):
             mgr = self._init_mgr(
@@ -545,10 +545,10 @@ def _is_homogeneous_type(self):
         ...    "B": np.array([1, 2], dtype=np.int64)})._is_homogeneous_type
         False
         """
-        if self._data.any_extension_types:
-            return len({block.dtype for block in self._data.blocks}) == 1
+        if self._mgr.any_extension_types:
+            return len({block.dtype for block in self._mgr.blocks}) == 1
         else:
-            return not self._data.is_mixed_type
+            return not self._mgr.is_mixed_type
 
     # ----------------------------------------------------------------------
     # Rendering Methods
@@ -2521,7 +2521,7 @@ def _sizeof_fmt(num, size_qualifier):
             else:
                 _verbose_repr()
 
-        counts = self._data.get_dtype_counts()
+        counts = self._mgr.get_dtype_counts()
         dtypes = ["{k}({kk:d})".format(k=k[0], kk=k[1]) for k in sorted(counts.items())]
         lines.append("dtypes: {types}".format(types=", ".join(dtypes)))
 
@@ -2755,7 +2755,7 @@ def _unpickle_frame_compat(self, state):  # pragma: no cover
             columns = com._unpickle_array(cols)
 
         index = com._unpickle_array(idx)
-        self._data = self._init_dict(series, index, columns, None)
+        self._mgr = self._init_dict(series, index, columns, None)
 
     def _unpickle_matrix_compat(self, state):  # pragma: no cover
         # old unpickling
@@ -2772,7 +2772,7 @@ def _unpickle_matrix_compat(self, state):  # pragma: no cover
 
             dm = dm.join(objects)
 
-        self._data = dm._data
+        self._mgr = dm._mgr
 
     # ----------------------------------------------------------------------
     # Getting and setting elements
@@ -2905,7 +2905,7 @@ def _ixs(self, i, axis=0):
                     result = self.take(i, axis=axis)
                     copy = True
                 else:
-                    new_values = self._data.fast_xs(i)
+                    new_values = self._mgr.fast_xs(i)
                     if is_scalar(new_values):
                         return new_values
 
@@ -2939,7 +2939,7 @@ def _ixs(self, i, axis=0):
                 # as the index (iow a not found value), iget returns
                 # a 0-len ndarray. This is effectively catching
                 # a numpy error (as numpy should really raise)
-                values = self._data.iget(i)
+                values = self._mgr.iget(i)
 
                 if index_len and not len(values):
                     values = np.array([np.nan] * index_len, dtype=object)
@@ -3538,7 +3538,7 @@ def _ensure_valid_index(self, value):
                     "Series"
                 )
 
-            self._data = self._data.reindex_axis(
+            self._mgr = self._mgr.reindex_axis(
                 value.index.copy(), axis=1, fill_value=np.nan
             )
 
@@ -3581,7 +3581,7 @@ def insert(self, loc, column, value, allow_duplicates=False):
         """
         self._ensure_valid_index(value)
         value = self._sanitize_column(column, value, broadcast=False)
-        self._data.insert(loc, column, value, allow_duplicates=allow_duplicates)
+        self._mgr.insert(loc, column, value, allow_duplicates=allow_duplicates)
 
     def assign(self, **kwargs):
         r"""
@@ -3780,7 +3780,7 @@ def reindexer(value):
     @property
     def _series(self):
         return {
-            item: Series(self._data.iget(idx), index=self.index, name=item)
+            item: Series(self._mgr.iget(idx), index=self.index, name=item)
             for idx, item in enumerate(self.columns)
         }
 
@@ -4903,7 +4903,7 @@ def drop_duplicates(self, subset=None, keep="first", inplace=False):
 
         if inplace:
             inds, = (-duplicated)._ndarray_values.nonzero()
-            new_data = self._data.take(inds)
+            new_data = self._mgr.take(inds)
             self._update_inplace(new_data)
         else:
             return self[-duplicated]
@@ -5007,7 +5007,7 @@ def sort_values(
                 k, kind=kind, ascending=ascending, na_position=na_position
             )
 
-        new_data = self._data.take(
+        new_data = self._mgr.take(
             indexer, axis=self._get_block_manager_axis(axis), verify=False
         )
 
@@ -5084,7 +5084,7 @@ def sort_index(
             )
 
         baxis = self._get_block_manager_axis(axis)
-        new_data = self._data.take(indexer, axis=baxis, verify=False)
+        new_data = self._mgr.take(indexer, axis=baxis, verify=False)
 
         # reconstruct axis if needed
         new_data.axes[baxis] = new_data.axes[baxis]._sort_levels_monotonic()
@@ -6527,7 +6527,7 @@ def diff(self, periods=1, axis=0):
         5  NaN  NaN   NaN
         """
         bm_axis = self._get_block_manager_axis(axis)
-        new_data = self._data.diff(n=periods, axis=bm_axis)
+        new_data = self._mgr.diff(n=periods, axis=bm_axis)
         return self._constructor(new_data)
 
     # ----------------------------------------------------------------------
@@ -7754,7 +7754,7 @@ def count(self, axis=0, level=None, numeric_only=False):
         if len(frame._get_axis(axis)) == 0:
             result = Series(0, index=frame._get_agg_axis(axis))
         else:
-            if frame._is_mixed_type or frame._data.any_extension_types:
+            if frame._is_mixed_type or frame._mgr.any_extension_types:
                 # the or any_extension_types is really only hit for single-
                 # column frames with an extension array
                 result = notna(frame).sum(axis=axis)
@@ -8209,7 +8209,7 @@ def quantile(self, q=0.5, axis=0, numeric_only=True, interpolation="linear"):
         if is_transposed:
             data = data.T
 
-        result = data._data.quantile(
+        result = data._mgr.quantile(
             qs=q, axis=1, interpolation=interpolation, transposed=is_transposed
         )
 
@@ -8243,7 +8243,7 @@ def to_timestamp(self, freq=None, how="start", axis=0, copy=True):
         -------
         DataFrame with DatetimeIndex
         """
-        new_data = self._data
+        new_data = self._mgr
         if copy:
             new_data = new_data.copy()
 
@@ -8275,7 +8275,7 @@ def to_period(self, freq=None, axis=0, copy=True):
         -------
         TimeSeries with PeriodIndex
         """
-        new_data = self._data
+        new_data = self._mgr
         if copy:
             new_data = new_data.copy()