From 155639b16b5006cec1be3dd7ec117e929f1e2ceb Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Mon, 22 Jul 2019 15:46:07 -0700
Subject: [PATCH 1/8] Fix fields functions with readonly data, vaex#357

---
 pandas/_libs/tslibs/fields.pyx     | 10 +++++-----
 pandas/tests/tslibs/test_fields.py | 29 +++++++++++++++++++++++++++++
 2 files changed, 34 insertions(+), 5 deletions(-)
 create mode 100644 pandas/tests/tslibs/test_fields.py

diff --git a/pandas/_libs/tslibs/fields.pyx b/pandas/_libs/tslibs/fields.pyx
index 2a41b5ff2339c..2ed85595f7e3a 100644
--- a/pandas/_libs/tslibs/fields.pyx
+++ b/pandas/_libs/tslibs/fields.pyx
@@ -45,7 +45,7 @@ def get_time_micros(ndarray[int64_t] dtindex):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def build_field_sarray(int64_t[:] dtindex):
+def build_field_sarray(const int64_t[:] dtindex):
     """
     Datetime as int64 representation to a structured array of fields
     """
@@ -87,7 +87,7 @@ def build_field_sarray(int64_t[:] dtindex):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_date_name_field(int64_t[:] dtindex, object field, object locale=None):
+def get_date_name_field(const int64_t[:] dtindex, object field, object locale=None):
     """
     Given a int64-based datetime index, return array of strings of date
     name based on requested field (e.g. weekday_name)
@@ -137,7 +137,7 @@ def get_date_name_field(int64_t[:] dtindex, object field, object locale=None):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_start_end_field(int64_t[:] dtindex, object field,
+def get_start_end_field(const int64_t[:] dtindex, object field,
                         object freqstr=None, int month_kw=12):
     """
     Given an int64-based datetime index return array of indicators
@@ -380,7 +380,7 @@ def get_start_end_field(int64_t[:] dtindex, object field,
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_date_field(int64_t[:] dtindex, object field):
+def get_date_field(const int64_t[:] dtindex, object field):
     """
     Given a int64-based datetime index, extract the year, month, etc.,
     field and return an array of these values.
@@ -542,7 +542,7 @@ def get_date_field(int64_t[:] dtindex, object field):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_timedelta_field(int64_t[:] tdindex, object field):
+def get_timedelta_field(const int64_t[:] tdindex, object field):
     """
     Given a int64-based timedelta index, extract the days, hrs, sec.,
     field and return an array of these values.
diff --git a/pandas/tests/tslibs/test_fields.py b/pandas/tests/tslibs/test_fields.py
new file mode 100644
index 0000000000000..7a01d1bca568d
--- /dev/null
+++ b/pandas/tests/tslibs/test_fields.py
@@ -0,0 +1,29 @@
+import numpy as np
+
+from pandas._libs.tslibs import fields
+import pandas.util.testing as tm
+
+
+def test_fields_readonly():
+	# https://github.com/vaexio/vaex/issues/357
+	#  fields functions should't raise when we pass read-only data
+	dtindex = np.arange(5, dtype=np.int64) * 10**9 * 3600 * 24 * 32
+	dtindex.flags.writeable = False
+
+	result = fields.get_date_name_field(dtindex, "month_name")
+	expected = np.array(['January', 'February', 'March', 'April', 'May'],
+				 	    dtype=np.object)
+	tm.assert_numpy_array_equal(result, expected)
+
+	result = fields.get_date_field(dtindex, "Y")
+	expected = np.array([1970, 1970, 1970, 1970, 1970], dtype=np.int32)
+	tm.assert_numpy_array_equal(result, expected)
+
+	result = fields.get_start_end_field(dtindex, "is_month_start", None)
+	expected = np.array([True, False, False, False, False], dtype=np.bool_)
+	tm.assert_numpy_array_equal(result, expected)
+
+	# treat dtindex as timedeltas for this next one
+	result = fields.get_timedelta_field(dtindex, "days")
+	expected = np.arange(5, dtype=np.int32) * 32
+	tm.assert_numpy_array_equal(result, expected)

From df6f11f614efdcf4384e43e4b131a8f354413b9d Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Mon, 22 Jul 2019 16:51:17 -0700
Subject: [PATCH 2/8] lint fixup

---
 pandas/tests/tslibs/test_fields.py | 37 +++++++++++++++---------------
 1 file changed, 19 insertions(+), 18 deletions(-)

diff --git a/pandas/tests/tslibs/test_fields.py b/pandas/tests/tslibs/test_fields.py
index 7a01d1bca568d..bd57758f31a75 100644
--- a/pandas/tests/tslibs/test_fields.py
+++ b/pandas/tests/tslibs/test_fields.py
@@ -5,25 +5,26 @@
 
 
 def test_fields_readonly():
-	# https://github.com/vaexio/vaex/issues/357
-	#  fields functions should't raise when we pass read-only data
-	dtindex = np.arange(5, dtype=np.int64) * 10**9 * 3600 * 24 * 32
-	dtindex.flags.writeable = False
+    # https://github.com/vaexio/vaex/issues/357
+    #  fields functions should't raise when we pass read-only data
+    dtindex = np.arange(5, dtype=np.int64) * 10 ** 9 * 3600 * 24 * 32
+    dtindex.flags.writeable = False
 
-	result = fields.get_date_name_field(dtindex, "month_name")
-	expected = np.array(['January', 'February', 'March', 'April', 'May'],
-				 	    dtype=np.object)
-	tm.assert_numpy_array_equal(result, expected)
+    result = fields.get_date_name_field(dtindex, "month_name")
+    expected = np.array(
+        ["January", "February", "March", "April", "May"], dtype=np.object
+    )
+    tm.assert_numpy_array_equal(result, expected)
 
-	result = fields.get_date_field(dtindex, "Y")
-	expected = np.array([1970, 1970, 1970, 1970, 1970], dtype=np.int32)
-	tm.assert_numpy_array_equal(result, expected)
+    result = fields.get_date_field(dtindex, "Y")
+    expected = np.array([1970, 1970, 1970, 1970, 1970], dtype=np.int32)
+    tm.assert_numpy_array_equal(result, expected)
 
-	result = fields.get_start_end_field(dtindex, "is_month_start", None)
-	expected = np.array([True, False, False, False, False], dtype=np.bool_)
-	tm.assert_numpy_array_equal(result, expected)
+    result = fields.get_start_end_field(dtindex, "is_month_start", None)
+    expected = np.array([True, False, False, False, False], dtype=np.bool_)
+    tm.assert_numpy_array_equal(result, expected)
 
-	# treat dtindex as timedeltas for this next one
-	result = fields.get_timedelta_field(dtindex, "days")
-	expected = np.arange(5, dtype=np.int32) * 32
-	tm.assert_numpy_array_equal(result, expected)
+    # treat dtindex as timedeltas for this next one
+    result = fields.get_timedelta_field(dtindex, "days")
+    expected = np.arange(5, dtype=np.int32) * 32
+    tm.assert_numpy_array_equal(result, expected)

From 74d8270f0a6dc501e6270cf4629224ca20b3968a Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Mon, 22 Jul 2019 19:40:42 -0700
Subject: [PATCH 3/8] cleanups

---
 pandas/_libs/missing.pyx        |  7 ++-----
 pandas/core/generic.py          | 14 +++++++-------
 pandas/core/groupby/generic.py  | 17 ++++++++---------
 pandas/core/internals/blocks.py |  3 +--
 pandas/core/reshape/pivot.py    |  2 --
 pandas/core/sorting.py          | 10 +++++-----
 pandas/core/window.py           |  6 ++----
 7 files changed, 25 insertions(+), 34 deletions(-)

diff --git a/pandas/_libs/missing.pyx b/pandas/_libs/missing.pyx
index 1d756115ebd5a..052b081988c9e 100644
--- a/pandas/_libs/missing.pyx
+++ b/pandas/_libs/missing.pyx
@@ -80,11 +80,8 @@ cpdef bint checknull_old(object val):
 
 
 cdef inline bint _check_none_nan_inf_neginf(object val):
-    try:
-        return val is None or (isinstance(val, float) and
-                               (val != val or val == INF or val == NEGINF))
-    except ValueError:
-        return False
+    return val is None or (isinstance(val, float) and
+                           (val != val or val == INF or val == NEGINF))
 
 
 @cython.wraparound(False)
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 0afd42e406c1f..d9cd6b67787c3 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -4893,12 +4893,12 @@ def sample(
         if weights is not None:
 
             # If a series, align with frame
-            if isinstance(weights, pd.Series):
+            if isinstance(weights, ABCSeries):
                 weights = weights.reindex(self.axes[axis])
 
             # Strings acceptable if a dataframe and axis = 0
             if isinstance(weights, str):
-                if isinstance(self, pd.DataFrame):
+                if isinstance(self, ABCDataFrame):
                     if axis == 0:
                         try:
                             weights = self[weights]
@@ -6629,7 +6629,7 @@ def replace(
                 to_replace = [to_replace]
 
             if isinstance(to_replace, (tuple, list)):
-                if isinstance(self, pd.DataFrame):
+                if isinstance(self, ABCDataFrame):
                     return self.apply(
                         _single_replace, args=(to_replace, method, inplace, limit)
                     )
@@ -7422,7 +7422,7 @@ def _clip_with_one_bound(self, threshold, method, axis, inplace):
         # be transformed to NDFrame from other array like structure.
         if (not isinstance(threshold, ABCSeries)) and is_list_like(threshold):
             if isinstance(self, ABCSeries):
-                threshold = pd.Series(threshold, index=self.index)
+                threshold = self._constructor(threshold, index=self.index)
             else:
                 threshold = _align_method_FRAME(self, threshold, axis)
         return self.where(subset, threshold, axis=axis, inplace=inplace)
@@ -7511,9 +7511,9 @@ def clip(self, lower=None, upper=None, axis=None, inplace=False, *args, **kwargs
         # so ignore
         # GH 19992
         # numpy doesn't drop a list-like bound containing NaN
-        if not is_list_like(lower) and np.any(pd.isnull(lower)):
+        if not is_list_like(lower) and np.any(pd.isna(lower)):
             lower = None
-        if not is_list_like(upper) and np.any(pd.isnull(upper)):
+        if not is_list_like(upper) and np.any(pd.isna(upper)):
             upper = None
 
         # GH 2747 (arguments were reversed)
@@ -8986,7 +8986,7 @@ def _where(
 
         msg = "Boolean array expected for the condition, not {dtype}"
 
-        if not isinstance(cond, pd.DataFrame):
+        if not isinstance(cond, ABCDataFrame):
             # This is a single-dimensional object.
             if not is_bool_dtype(cond):
                 raise ValueError(msg.format(dtype=cond.dtype))
diff --git a/pandas/core/groupby/generic.py b/pandas/core/groupby/generic.py
index 5b9cec6903749..8ea9124b0e98e 100644
--- a/pandas/core/groupby/generic.py
+++ b/pandas/core/groupby/generic.py
@@ -35,7 +35,7 @@
     is_object_dtype,
     is_scalar,
 )
-from pandas.core.dtypes.missing import isna, notna
+from pandas.core.dtypes.missing import _isna_ndarraylike, isna, notna
 
 from pandas._typing import FrameOrSeries
 import pandas.core.algorithms as algorithms
@@ -45,7 +45,7 @@
 from pandas.core.generic import ABCDataFrame, ABCSeries, NDFrame, _shared_docs
 from pandas.core.groupby import base
 from pandas.core.groupby.groupby import GroupBy, _apply_docs, _transform_template
-from pandas.core.index import Index, MultiIndex
+from pandas.core.index import Index, MultiIndex, _all_indexes_same
 import pandas.core.indexes.base as ibase
 from pandas.core.internals import BlockManager, make_block
 from pandas.core.series import Series
@@ -53,6 +53,8 @@
 
 from pandas.plotting import boxplot_frame_groupby
 
+from .groupby import groupby
+
 NamedAgg = namedtuple("NamedAgg", ["column", "aggfunc"])
 # TODO(typing) the return value on this callable should be any *scalar*.
 AggScalar = Union[str, Callable[..., Any]]
@@ -162,8 +164,6 @@ def _cython_agg_blocks(self, how, alt=None, numeric_only=True, min_count=-1):
                     continue
 
                 # call our grouper again with only this block
-                from pandas.core.groupby.groupby import groupby
-
                 obj = self.obj[data.items[locs]]
                 s = groupby(obj, self.grouper)
                 try:
@@ -348,8 +348,6 @@ def _decide_output_index(self, output, labels):
         return output_keys
 
     def _wrap_applied_output(self, keys, values, not_indexed_same=False):
-        from pandas.core.index import _all_indexes_same
-
         if len(keys) == 0:
             return DataFrame(index=keys)
 
@@ -1590,13 +1588,14 @@ def count(self):
         DataFrame
             Count of values within each group.
         """
-        from pandas.core.dtypes.missing import _isna_ndarraylike as _isna
-
         data, _ = self._get_data_to_aggregate()
         ids, _, ngroups = self.grouper.group_info
         mask = ids != -1
 
-        val = ((mask & ~_isna(np.atleast_2d(blk.get_values()))) for blk in data.blocks)
+        val = (
+            (mask & ~_isna_ndarraylike(np.atleast_2d(blk.get_values())))
+            for blk in data.blocks
+        )
         loc = (blk.mgr_locs for blk in data.blocks)
 
         counter = partial(lib.count_level_2d, labels=ids, max_bin=ngroups, axis=1)
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index f0e7893435f2b..82fee5c9e36d7 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -75,6 +75,7 @@
     PandasDtype,
     TimedeltaArray,
 )
+from pandas.core.arrays.categorical import _maybe_to_categorical
 from pandas.core.base import PandasObject
 import pandas.core.common as com
 from pandas.core.indexers import (
@@ -3085,8 +3086,6 @@ class CategoricalBlock(ExtensionBlock):
     _concatenator = staticmethod(concat_categorical)
 
     def __init__(self, values, placement, ndim=None):
-        from pandas.core.arrays.categorical import _maybe_to_categorical
-
         # coerce to categorical if we can
         super().__init__(_maybe_to_categorical(values), placement=placement, ndim=ndim)
 
diff --git a/pandas/core/reshape/pivot.py b/pandas/core/reshape/pivot.py
index 2bdef766a3434..79716520f6654 100644
--- a/pandas/core/reshape/pivot.py
+++ b/pandas/core/reshape/pivot.py
@@ -127,8 +127,6 @@ def pivot_table(
         table = agged.unstack(to_unstack)
 
     if not dropna:
-        from pandas import MultiIndex
-
         if table.index.nlevels > 1:
             m = MultiIndex.from_arrays(
                 cartesian_product(table.index.levels), names=table.index.names
diff --git a/pandas/core/sorting.py b/pandas/core/sorting.py
index 5f3ed87424d0e..1ab6c792c6402 100644
--- a/pandas/core/sorting.py
+++ b/pandas/core/sorting.py
@@ -202,19 +202,19 @@ def lexsort_indexer(keys, orders=None, na_position="last"):
 
         # we are already a Categorical
         if is_categorical_dtype(key):
-            c = key
+            cat = key
 
         # create the Categorical
         else:
-            c = Categorical(key, ordered=True)
+            cat = Categorical(key, ordered=True)
 
         if na_position not in ["last", "first"]:
             raise ValueError("invalid na_position: {!r}".format(na_position))
 
-        n = len(c.categories)
-        codes = c.codes.copy()
+        n = len(cat.categories)
+        codes = cat.codes.copy()
 
-        mask = c.codes == -1
+        mask = cat.codes == -1
         if order:  # ascending
             if na_position == "last":
                 codes = np.where(mask, n, codes)
diff --git a/pandas/core/window.py b/pandas/core/window.py
index 5098ab3c7220f..14d109ccf6a9c 100644
--- a/pandas/core/window.py
+++ b/pandas/core/window.py
@@ -40,6 +40,7 @@
 import pandas.core.common as com
 from pandas.core.generic import _shared_docs
 from pandas.core.groupby.base import GroupByMixin
+from pandas.core.index import Index, MultiIndex, ensure_index
 
 _shared_docs = dict(**_shared_docs)
 _doc_template = """
@@ -281,7 +282,6 @@ def _wrap_results(self, results, blocks, obj, exclude=None) -> FrameOrSeries:
         """
 
         from pandas import Series, concat
-        from pandas.core.index import ensure_index
 
         final = []
         for result, block in zip(results, blocks):
@@ -1691,8 +1691,6 @@ def _on(self):
         if self.on is None:
             return self.obj.index
         elif isinstance(self.obj, ABCDataFrame) and self.on in self.obj.columns:
-            from pandas import Index
-
             return Index(self.obj[self.on])
         else:
             raise ValueError(
@@ -2670,7 +2668,7 @@ def dataframe_from_int_dict(data, frame_template):
                                 *_prep_binary(arg1.iloc[:, i], arg2.iloc[:, j])
                             )
 
-                from pandas import MultiIndex, concat
+                from pandas import concat
 
                 result_index = arg1.index.union(arg2.index)
                 if len(result_index):

From 714b878f1ac105c49e4028ed41db6980d60180e2 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Tue, 23 Jul 2019 15:14:51 -0700
Subject: [PATCH 4/8] revert things that are in deps2

---
 pandas/_libs/tslibs/fields.pyx     | 10 +++++-----
 pandas/tests/tslibs/test_fields.py | 30 ------------------------------
 2 files changed, 5 insertions(+), 35 deletions(-)
 delete mode 100644 pandas/tests/tslibs/test_fields.py

diff --git a/pandas/_libs/tslibs/fields.pyx b/pandas/_libs/tslibs/fields.pyx
index 2ed85595f7e3a..2a41b5ff2339c 100644
--- a/pandas/_libs/tslibs/fields.pyx
+++ b/pandas/_libs/tslibs/fields.pyx
@@ -45,7 +45,7 @@ def get_time_micros(ndarray[int64_t] dtindex):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def build_field_sarray(const int64_t[:] dtindex):
+def build_field_sarray(int64_t[:] dtindex):
     """
     Datetime as int64 representation to a structured array of fields
     """
@@ -87,7 +87,7 @@ def build_field_sarray(const int64_t[:] dtindex):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_date_name_field(const int64_t[:] dtindex, object field, object locale=None):
+def get_date_name_field(int64_t[:] dtindex, object field, object locale=None):
     """
     Given a int64-based datetime index, return array of strings of date
     name based on requested field (e.g. weekday_name)
@@ -137,7 +137,7 @@ def get_date_name_field(const int64_t[:] dtindex, object field, object locale=No
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_start_end_field(const int64_t[:] dtindex, object field,
+def get_start_end_field(int64_t[:] dtindex, object field,
                         object freqstr=None, int month_kw=12):
     """
     Given an int64-based datetime index return array of indicators
@@ -380,7 +380,7 @@ def get_start_end_field(const int64_t[:] dtindex, object field,
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_date_field(const int64_t[:] dtindex, object field):
+def get_date_field(int64_t[:] dtindex, object field):
     """
     Given a int64-based datetime index, extract the year, month, etc.,
     field and return an array of these values.
@@ -542,7 +542,7 @@ def get_date_field(const int64_t[:] dtindex, object field):
 
 @cython.wraparound(False)
 @cython.boundscheck(False)
-def get_timedelta_field(const int64_t[:] tdindex, object field):
+def get_timedelta_field(int64_t[:] tdindex, object field):
     """
     Given a int64-based timedelta index, extract the days, hrs, sec.,
     field and return an array of these values.
diff --git a/pandas/tests/tslibs/test_fields.py b/pandas/tests/tslibs/test_fields.py
deleted file mode 100644
index bd57758f31a75..0000000000000
--- a/pandas/tests/tslibs/test_fields.py
+++ /dev/null
@@ -1,30 +0,0 @@
-import numpy as np
-
-from pandas._libs.tslibs import fields
-import pandas.util.testing as tm
-
-
-def test_fields_readonly():
-    # https://github.com/vaexio/vaex/issues/357
-    #  fields functions should't raise when we pass read-only data
-    dtindex = np.arange(5, dtype=np.int64) * 10 ** 9 * 3600 * 24 * 32
-    dtindex.flags.writeable = False
-
-    result = fields.get_date_name_field(dtindex, "month_name")
-    expected = np.array(
-        ["January", "February", "March", "April", "May"], dtype=np.object
-    )
-    tm.assert_numpy_array_equal(result, expected)
-
-    result = fields.get_date_field(dtindex, "Y")
-    expected = np.array([1970, 1970, 1970, 1970, 1970], dtype=np.int32)
-    tm.assert_numpy_array_equal(result, expected)
-
-    result = fields.get_start_end_field(dtindex, "is_month_start", None)
-    expected = np.array([True, False, False, False, False], dtype=np.bool_)
-    tm.assert_numpy_array_equal(result, expected)
-
-    # treat dtindex as timedeltas for this next one
-    result = fields.get_timedelta_field(dtindex, "days")
-    expected = np.arange(5, dtype=np.int32) * 32
-    tm.assert_numpy_array_equal(result, expected)

From bf29a7c93fb8db93ae81822b561f564bcafbf6c2 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Tue, 23 Jul 2019 18:00:30 -0700
Subject: [PATCH 5/8] remove maybe_to_categorical

---
 pandas/core/arrays/categorical.py | 13 -------------
 pandas/core/internals/blocks.py   |  5 +++--
 2 files changed, 3 insertions(+), 15 deletions(-)

diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 6200cd14663f8..8adbc0fbd6d35 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -166,19 +166,6 @@ def f(self, other):
     return f
 
 
-def _maybe_to_categorical(array):
-    """
-    Coerce to a categorical if a series is given.
-
-    Internal use ONLY.
-    """
-    if isinstance(array, (ABCSeries, ABCCategoricalIndex)):
-        return array._values
-    elif isinstance(array, np.ndarray):
-        return Categorical(array)
-    return array
-
-
 def contains(cat, key, container):
     """
     Helper for membership check for ``key`` in ``cat``.
diff --git a/pandas/core/internals/blocks.py b/pandas/core/internals/blocks.py
index 82fee5c9e36d7..29b1942e9d9b4 100644
--- a/pandas/core/internals/blocks.py
+++ b/pandas/core/internals/blocks.py
@@ -75,7 +75,6 @@
     PandasDtype,
     TimedeltaArray,
 )
-from pandas.core.arrays.categorical import _maybe_to_categorical
 from pandas.core.base import PandasObject
 import pandas.core.common as com
 from pandas.core.indexers import (
@@ -3087,7 +3086,9 @@ class CategoricalBlock(ExtensionBlock):
 
     def __init__(self, values, placement, ndim=None):
         # coerce to categorical if we can
-        super().__init__(_maybe_to_categorical(values), placement=placement, ndim=ndim)
+        values = extract_array(values)
+        assert isinstance(values, Categorical), type(values)
+        super().__init__(values, placement=placement, ndim=ndim)
 
     @property
     def _holder(self):

From 5bc46ec8b22fe7890d44e350890dd3c1e1ea7094 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Tue, 23 Jul 2019 20:38:44 -0700
Subject: [PATCH 6/8] cln

---
 pandas/core/arrays/categorical.py | 18 ------------------
 pandas/core/arrays/sparse.py      | 13 -------------
 pandas/core/sparse/series.py      |  5 +++++
 3 files changed, 5 insertions(+), 31 deletions(-)

diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 8adbc0fbd6d35..08e247da4cb3b 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -39,7 +39,6 @@
 )
 from pandas.core.dtypes.dtypes import CategoricalDtype
 from pandas.core.dtypes.generic import (
-    ABCCategoricalIndex,
     ABCDataFrame,
     ABCIndexClass,
     ABCSeries,
@@ -1975,23 +1974,6 @@ def take_nd(self, indexer, allow_fill=None, fill_value=None):
 
     take = take_nd
 
-    def _slice(self, slicer):
-        """
-        Return a slice of myself.
-
-        For internal compatibility with numpy arrays.
-        """
-
-        # only allow 1 dimensional slicing, but can
-        # in a 2-d case be passd (slice(None),....)
-        if isinstance(slicer, tuple) and len(slicer) == 2:
-            if not com.is_null_slice(slicer[0]):
-                raise AssertionError("invalid slicing for a 1-ndim " "categorical")
-            slicer = slicer[1]
-
-        codes = self._codes[slicer]
-        return self._constructor(values=codes, dtype=self.dtype, fastpath=True)
-
     def __len__(self):
         """
         The length of this Categorical.
diff --git a/pandas/core/arrays/sparse.py b/pandas/core/arrays/sparse.py
index 9376b49112f6f..ee3652a211e31 100644
--- a/pandas/core/arrays/sparse.py
+++ b/pandas/core/arrays/sparse.py
@@ -601,10 +601,6 @@ def __init__(
         dtype=None,
         copy=False,
     ):
-        from pandas.core.internals import SingleBlockManager
-
-        if isinstance(data, SingleBlockManager):
-            data = data.internal_values()
 
         if fill_value is None and isinstance(dtype, SparseDtype):
             fill_value = dtype.fill_value
@@ -1859,15 +1855,6 @@ def _formatter(self, boxed=False):
 SparseArray._add_unary_ops()
 
 
-def _maybe_to_dense(obj):
-    """
-    try to convert to dense
-    """
-    if hasattr(obj, "to_dense"):
-        return obj.to_dense()
-    return obj
-
-
 def make_sparse(arr, kind="block", fill_value=None, dtype=None, copy=False):
     """
     Convert ndarray to sparse format
diff --git a/pandas/core/sparse/series.py b/pandas/core/sparse/series.py
index f5d39c47150a2..0bd20e75d17ec 100644
--- a/pandas/core/sparse/series.py
+++ b/pandas/core/sparse/series.py
@@ -114,6 +114,11 @@ def __init__(
         elif is_scalar(data) and index is not None:
             data = np.full(len(index), fill_value=data)
 
+        if isinstance(data, SingleBlockManager):
+            # SparseArray doesn't accept SingleBlockManager
+            index = data.index
+            data = data.blocks[0].values
+
         super().__init__(
             SparseArray(
                 data,

From 46898efc77893c030026c792e5f3a664eee42642 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Tue, 23 Jul 2019 20:40:25 -0700
Subject: [PATCH 7/8] isort

---
 pandas/core/arrays/categorical.py | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/pandas/core/arrays/categorical.py b/pandas/core/arrays/categorical.py
index 08e247da4cb3b..dd71daab9d4c5 100644
--- a/pandas/core/arrays/categorical.py
+++ b/pandas/core/arrays/categorical.py
@@ -38,11 +38,7 @@
     is_timedelta64_dtype,
 )
 from pandas.core.dtypes.dtypes import CategoricalDtype
-from pandas.core.dtypes.generic import (
-    ABCDataFrame,
-    ABCIndexClass,
-    ABCSeries,
-)
+from pandas.core.dtypes.generic import ABCDataFrame, ABCIndexClass, ABCSeries
 from pandas.core.dtypes.inference import is_hashable
 from pandas.core.dtypes.missing import isna, notna
 

From 925ce8103c34e26e828b9f309531d3245db268cd Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Wed, 24 Jul 2019 06:27:15 -0700
Subject: [PATCH 8/8] comments

---
 pandas/core/generic.py         | 4 ++--
 pandas/core/groupby/generic.py | 9 ++++++---
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 7291f113dc6f2..1a854af52c20e 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -7510,9 +7510,9 @@ def clip(self, lower=None, upper=None, axis=None, inplace=False, *args, **kwargs
         # so ignore
         # GH 19992
         # numpy doesn't drop a list-like bound containing NaN
-        if not is_list_like(lower) and np.any(pd.isna(lower)):
+        if not is_list_like(lower) and np.any(isna(lower)):
             lower = None
-        if not is_list_like(upper) and np.any(pd.isna(upper)):
+        if not is_list_like(upper) and np.any(isna(upper)):
             upper = None
 
         # GH 2747 (arguments were reversed)
diff --git a/pandas/core/groupby/generic.py b/pandas/core/groupby/generic.py
index 8ea9124b0e98e..b886b7e305ed0 100644
--- a/pandas/core/groupby/generic.py
+++ b/pandas/core/groupby/generic.py
@@ -44,7 +44,12 @@
 from pandas.core.frame import DataFrame
 from pandas.core.generic import ABCDataFrame, ABCSeries, NDFrame, _shared_docs
 from pandas.core.groupby import base
-from pandas.core.groupby.groupby import GroupBy, _apply_docs, _transform_template
+from pandas.core.groupby.groupby import (
+    GroupBy,
+    _apply_docs,
+    _transform_template,
+    groupby,
+)
 from pandas.core.index import Index, MultiIndex, _all_indexes_same
 import pandas.core.indexes.base as ibase
 from pandas.core.internals import BlockManager, make_block
@@ -53,8 +58,6 @@
 
 from pandas.plotting import boxplot_frame_groupby
 
-from .groupby import groupby
-
 NamedAgg = namedtuple("NamedAgg", ["column", "aggfunc"])
 # TODO(typing) the return value on this callable should be any *scalar*.
 AggScalar = Union[str, Callable[..., Any]]