CLN: avoid runtime imports (#27461)

Author: jbrockmendel

pandas/_libs/lib.pyx

@@ -157,13 +157,13 @@ def is_scalar(val: object) -> bool:
 
     return (cnp.PyArray_IsAnyScalar(val)
             # PyArray_IsAnyScalar is always False for bytearrays on Py3
-            or isinstance(val, (Fraction, Number))
-            # We differ from numpy, which claims that None is not scalar;
-            # see np.isscalar
-            or val is None
             or PyDate_Check(val)
             or PyDelta_Check(val)
             or PyTime_Check(val)
+            # We differ from numpy, which claims that None is not scalar;
+            # see np.isscalar
+            or val is None
+            or isinstance(val, (Fraction, Number))
             or util.is_period_object(val)
             or is_decimal(val)
             or is_interval(val)
@@ -1192,7 +1192,9 @@ def infer_dtype(value: object, skipna: object=None) -> str:
         # e.g. categoricals
         try:
             values = getattr(value, '_values', getattr(value, 'values', value))
-        except:
+        except TypeError:
+            # This gets hit if we have an EA, since cython expects `values`
+            #  to be an ndarray
             value = _try_infer_map(value)
             if value is not None:
                 return value
@@ -1208,8 +1210,6 @@ def infer_dtype(value: object, skipna: object=None) -> str:
             construct_1d_object_array_from_listlike)
         values = construct_1d_object_array_from_listlike(value)
 
-    values = getattr(values, 'values', values)
-
     # make contiguous
     values = values.ravel()
 

pandas/compat/pickle_compat.py

@@ -6,7 +6,6 @@
 import pickle as pkl
 import sys
 
-import pandas  # noqa
 from pandas import Index
 
 

pandas/core/algorithms.py

@@ -1977,12 +1977,6 @@ def diff(arr, n, axis=0):
             out_arr[res_indexer] = arr[res_indexer] - arr[lag_indexer]
 
     if is_timedelta:
-        from pandas import TimedeltaIndex
-
-        out_arr = (
-            TimedeltaIndex(out_arr.ravel().astype("int64"))
-            .asi8.reshape(out_arr.shape)
-            .astype("timedelta64[ns]")
-        )
+        out_arr = out_arr.astype("int64").view("timedelta64[ns]")
 
     return out_arr

pandas/core/arrays/array_.py

@@ -212,7 +212,6 @@ def array(
     """
     from pandas.core.arrays import (
         period_array,
-        ExtensionArray,
         IntervalArray,
         PandasArray,
         DatetimeArray,
@@ -226,7 +225,7 @@ def array(
 
     data = extract_array(data, extract_numpy=True)
 
-    if dtype is None and isinstance(data, ExtensionArray):
+    if dtype is None and isinstance(data, ABCExtensionArray):
         dtype = data.dtype
 
     # this returns None for not-found dtypes.

pandas/core/arrays/categorical.py

@@ -6,7 +6,7 @@
 
 from pandas._config import get_option
 
-from pandas._libs import algos as libalgos, lib
+from pandas._libs import algos as libalgos, hashtable as htable, lib
 from pandas.compat.numpy import function as nv
 from pandas.util._decorators import (
     Appender,
@@ -50,7 +50,14 @@
 from pandas.core import ops
 from pandas.core.accessor import PandasDelegate, delegate_names
 import pandas.core.algorithms as algorithms
-from pandas.core.algorithms import factorize, take, take_1d, unique1d
+from pandas.core.algorithms import (
+    _get_data_algo,
+    _hashtables,
+    factorize,
+    take,
+    take_1d,
+    unique1d,
+)
 from pandas.core.base import NoNewAttributesMixin, PandasObject, _shared_docs
 import pandas.core.common as com
 from pandas.core.missing import interpolate_2d
@@ -1527,9 +1534,7 @@ def value_counts(self, dropna=True):
         See Also
         --------
         Series.value_counts
-
         """
-        from numpy import bincount
         from pandas import Series, CategoricalIndex
 
         code, cat = self._codes, self.categories
@@ -1538,9 +1543,9 @@ def value_counts(self, dropna=True):
 
         if dropna or clean:
             obs = code if clean else code[mask]
-            count = bincount(obs, minlength=ncat or 0)
+            count = np.bincount(obs, minlength=ncat or 0)
         else:
-            count = bincount(np.where(mask, code, ncat))
+            count = np.bincount(np.where(mask, code, ncat))
             ix = np.append(ix, -1)
 
         ix = self._constructor(ix, dtype=self.dtype, fastpath=True)
@@ -2329,9 +2334,6 @@ def mode(self, dropna=True):
         -------
         modes : `Categorical` (sorted)
         """
-
-        import pandas._libs.hashtable as htable
-
         codes = self._codes
         if dropna:
             good = self._codes != -1
@@ -2671,8 +2673,6 @@ def _get_codes_for_values(values, categories):
     """
     utility routine to turn values into codes given the specified categories
     """
-    from pandas.core.algorithms import _get_data_algo, _hashtables
-
     dtype_equal = is_dtype_equal(values.dtype, categories.dtype)
 
     if dtype_equal:
@@ -2722,8 +2722,6 @@ def _recode_for_categories(codes, old_categories, new_categories):
     >>> _recode_for_categories(codes, old_cat, new_cat)
     array([ 1,  0,  0, -1])
     """
-    from pandas.core.algorithms import take_1d
-
     if len(old_categories) == 0:
         # All null anyway, so just retain the nulls
         return codes.copy()

pandas/core/arrays/datetimelike.py

@@ -34,7 +34,12 @@
     is_unsigned_integer_dtype,
     pandas_dtype,
 )
-from pandas.core.dtypes.generic import ABCDataFrame, ABCIndexClass, ABCSeries
+from pandas.core.dtypes.generic import (
+    ABCDataFrame,
+    ABCIndexClass,
+    ABCPeriodArray,
+    ABCSeries,
+)
 from pandas.core.dtypes.inference import is_array_like
 from pandas.core.dtypes.missing import is_valid_nat_for_dtype, isna
 
@@ -1679,11 +1684,10 @@ def _ensure_datetimelike_to_i8(other, to_utc=False):
     i8 1d array
     """
     from pandas import Index
-    from pandas.core.arrays import PeriodArray
 
     if lib.is_scalar(other) and isna(other):
         return iNaT
-    elif isinstance(other, (PeriodArray, ABCIndexClass, DatetimeLikeArrayMixin)):
+    elif isinstance(other, (ABCPeriodArray, ABCIndexClass, DatetimeLikeArrayMixin)):
         # convert tz if needed
         if getattr(other, "tz", None) is not None:
             if to_utc:

pandas/core/arrays/integer.py

@@ -25,6 +25,7 @@
 from pandas.core.dtypes.missing import isna, notna
 
 from pandas.core import nanops, ops
+from pandas.core.algorithms import take
 from pandas.core.arrays import ExtensionArray, ExtensionOpsMixin
 from pandas.core.tools.numeric import to_numeric
 
@@ -420,8 +421,6 @@ def __iter__(self):
                 yield self._data[i]
 
     def take(self, indexer, allow_fill=False, fill_value=None):
-        from pandas.api.extensions import take
-
         # we always fill with 1 internally
         # to avoid upcasting
         data_fill_value = 1 if isna(fill_value) else fill_value

pandas/core/arrays/interval.py

@@ -25,6 +25,7 @@
 from pandas.core.dtypes.dtypes import IntervalDtype
 from pandas.core.dtypes.generic import (
     ABCDatetimeIndex,
+    ABCIndexClass,
     ABCInterval,
     ABCIntervalIndex,
     ABCPeriodIndex,
@@ -35,7 +36,7 @@
 from pandas.core.arrays.base import ExtensionArray, _extension_array_shared_docs
 from pandas.core.arrays.categorical import Categorical
 import pandas.core.common as com
-from pandas.core.indexes.base import Index, ensure_index
+from pandas.core.indexes.base import ensure_index
 
 _VALID_CLOSED = {"left", "right", "both", "neither"}
 _interval_shared_docs = {}
@@ -510,7 +511,7 @@ def __getitem__(self, value):
         right = self.right[value]
 
         # scalar
-        if not isinstance(left, Index):
+        if not isinstance(left, ABCIndexClass):
             if isna(left):
                 return self._fill_value
             return Interval(left, right, self.closed)

pandas/core/arrays/numpy_.py

@@ -11,10 +11,11 @@
 from pandas.core.dtypes.dtypes import ExtensionDtype
 from pandas.core.dtypes.generic import ABCIndexClass, ABCSeries
 from pandas.core.dtypes.inference import is_array_like, is_list_like
+from pandas.core.dtypes.missing import isna
 
 from pandas import compat
 from pandas.core import nanops
-from pandas.core.algorithms import searchsorted
+from pandas.core.algorithms import searchsorted, take, unique
 from pandas.core.missing import backfill_1d, pad_1d
 
 from .base import ExtensionArray, ExtensionOpsMixin
@@ -249,8 +250,6 @@ def nbytes(self):
         return self._ndarray.nbytes
 
     def isna(self):
-        from pandas import isna
-
         return isna(self._ndarray)
 
     def fillna(self, value=None, method=None, limit=None):
@@ -281,8 +280,6 @@ def fillna(self, value=None, method=None, limit=None):
         return new_values
 
     def take(self, indices, allow_fill=False, fill_value=None):
-        from pandas.core.algorithms import take
-
         result = take(
             self._ndarray, indices, allow_fill=allow_fill, fill_value=fill_value
         )
@@ -298,8 +295,6 @@ def _values_for_factorize(self):
         return self._ndarray, -1
 
     def unique(self):
-        from pandas import unique
-
         return type(self)(unique(self._ndarray))
 
     # ------------------------------------------------------------------------

pandas/core/arrays/sparse.py

@@ -105,8 +105,6 @@ class SparseDtype(ExtensionDtype):
     _metadata = ("_dtype", "_fill_value", "_is_na_fill_value")
 
     def __init__(self, dtype: Dtype = np.float64, fill_value: Any = None) -> None:
-        from pandas.core.dtypes.missing import na_value_for_dtype
-        from pandas.core.dtypes.common import pandas_dtype, is_string_dtype, is_scalar
 
         if isinstance(dtype, type(self)):
             if fill_value is None:
@@ -178,20 +176,14 @@ def fill_value(self):
 
     @property
     def _is_na_fill_value(self):
-        from pandas.core.dtypes.missing import isna
-
         return isna(self.fill_value)
 
     @property
     def _is_numeric(self):
-        from pandas.core.dtypes.common import is_object_dtype
-
         return not is_object_dtype(self.subtype)
 
     @property
     def _is_boolean(self):
-        from pandas.core.dtypes.common import is_bool_dtype
-
         return is_bool_dtype(self.subtype)
 
     @property
@@ -928,8 +920,6 @@ def values(self):
         return self.to_dense()
 
     def isna(self):
-        from pandas import isna
-
         # If null fill value, we want SparseDtype[bool, true]
         # to preserve the same memory usage.
         dtype = SparseDtype(bool, self._null_fill_value)

pandas/core/base.py

@@ -32,6 +32,7 @@
 
 from pandas.core import algorithms, common as com
 from pandas.core.accessor import DirNamesMixin
+from pandas.core.algorithms import duplicated, unique1d, value_counts
 from pandas.core.arrays import ExtensionArray
 import pandas.core.nanops as nanops
 
@@ -1381,8 +1382,6 @@ def value_counts(
         1.0    1
         dtype: int64
         """
-        from pandas.core.algorithms import value_counts
-
         result = value_counts(
             self,
             sort=sort,
@@ -1400,8 +1399,6 @@ def unique(self):
 
             result = values.unique()
         else:
-            from pandas.core.algorithms import unique1d
-
             result = unique1d(values)
 
         return result
@@ -1631,8 +1628,6 @@ def drop_duplicates(self, keep="first", inplace=False):
             return result
 
     def duplicated(self, keep="first"):
-        from pandas.core.algorithms import duplicated
-
         if isinstance(self, ABCIndexClass):
             if self.is_unique:
                 return np.zeros(len(self), dtype=np.bool)

pandas/core/common.py

@@ -254,7 +254,6 @@ def asarray_tuplesafe(values, dtype=None):
 
     if result.ndim == 2:
         # Avoid building an array of arrays:
-        # TODO: verify whether any path hits this except #18819 (invalid)
         values = [tuple(x) for x in values]
         result = construct_1d_object_array_from_listlike(values)
 

pandas/core/dtypes/concat.py

@@ -361,9 +361,7 @@ def _maybe_unwrap(x):
             new_codes = np.concatenate(codes)
 
         if sort_categories and not ignore_order and ordered:
-            raise TypeError(
-                "Cannot use sort_categories=True with " "ordered Categoricals"
-            )
+            raise TypeError("Cannot use sort_categories=True with ordered Categoricals")
 
         if sort_categories and not categories.is_monotonic_increasing:
             categories = categories.sort_values()
@@ -386,7 +384,7 @@ def _maybe_unwrap(x):
     else:
         # ordered - to show a proper error message
         if all(c.ordered for c in to_union):
-            msg = "to union ordered Categoricals, " "all categories must be the same"
+            msg = "to union ordered Categoricals, all categories must be the same"
             raise TypeError(msg)
         else:
             raise TypeError("Categorical.ordered must be the same")

pandas/core/dtypes/dtypes.py

@@ -12,7 +12,7 @@
 from pandas.core.dtypes.generic import ABCCategoricalIndex, ABCDateOffset, ABCIndexClass
 
 from .base import ExtensionDtype
-from .inference import is_list_like
+from .inference import is_bool, is_list_like
 
 str_type = str
 
@@ -149,7 +149,7 @@ def __repr__(self) -> str_type:
         return str(self)
 
     def __hash__(self) -> int:
-        raise NotImplementedError("sub-classes should implement an __hash__ " "method")
+        raise NotImplementedError("sub-classes should implement an __hash__ method")
 
     def __getstate__(self) -> Dict[str_type, Any]:
         # pickle support; we don't want to pickle the cache
@@ -320,7 +320,7 @@ def _from_values_or_dtype(
                     raise ValueError(msg.format(dtype=dtype))
             elif categories is not None or ordered is not None:
                 raise ValueError(
-                    "Cannot specify `categories` or `ordered` " "together with `dtype`."
+                    "Cannot specify `categories` or `ordered` together with `dtype`."
                 )
         elif is_categorical(values):
             # If no "dtype" was passed, use the one from "values", but honor
@@ -490,8 +490,6 @@ def validate_ordered(ordered: OrderedType) -> None:
         TypeError
             If 'ordered' is not a boolean.
         """
-        from pandas.core.dtypes.common import is_bool
-
         if not is_bool(ordered):
             raise TypeError("'ordered' must either be 'True' or 'False'")