From 5506cc6b78cff164fccc7b19ec23ae205e28588f Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 09:31:53 -0700
Subject: [PATCH 01/12] mark indexing methods

---
 pandas/core/frame.py    | 15 +++++++++++++++
 pandas/core/generic.py  | 15 ++++++++++++---
 pandas/core/indexing.py |  6 ------
 pandas/core/series.py   | 17 +++++++++++++++++
 4 files changed, 44 insertions(+), 9 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 55a9eb6a0810a..6035f5c4461aa 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2751,6 +2751,7 @@ def transpose(self, *args, **kwargs):
     # ----------------------------------------------------------------------
     # Picklability
 
+    # TODO: can we get rid of these?
     # legacy pickle formats
     def _unpickle_frame_compat(self, state):  # pragma: no cover
         if len(state) == 2:  # pragma: no cover
@@ -2783,6 +2784,7 @@ def _unpickle_matrix_compat(self, state):  # pragma: no cover
     # ----------------------------------------------------------------------
     # Getting and setting elements
 
+    # indexing method
     def get_value(self, index, col, takeable=False):
         """
         Quickly retrieve single value at passed column and index.
@@ -2810,6 +2812,7 @@ def get_value(self, index, col, takeable=False):
         )
         return self._get_value(index, col, takeable=takeable)
 
+    # indexing method
     def _get_value(self, index, col, takeable=False):
 
         if takeable:
@@ -2837,6 +2840,7 @@ def _get_value(self, index, col, takeable=False):
 
     _get_value.__doc__ = get_value.__doc__
 
+    # indexing method
     def set_value(self, index, col, value, takeable=False):
         """
         Put single value at passed column and index.
@@ -2866,6 +2870,7 @@ def set_value(self, index, col, value, takeable=False):
         )
         return self._set_value(index, col, value, takeable=takeable)
 
+    # indexing method
     def _set_value(self, index, col, value, takeable=False):
         try:
             if takeable is True:
@@ -2889,6 +2894,7 @@ def _set_value(self, index, col, value, takeable=False):
 
     _set_value.__doc__ = set_value.__doc__
 
+    # indexing method
     def _ixs(self, i, axis=0):
         """
         Parameters
@@ -2956,6 +2962,7 @@ def _ixs(self, i, axis=0):
 
                 return result
 
+    # indexing method
     def __getitem__(self, key):
         key = lib.item_from_zerodim(key)
         key = com.apply_if_callable(key, self)
@@ -3011,6 +3018,7 @@ def __getitem__(self, key):
 
         return data
 
+    # indexing method
     def _getitem_bool_array(self, key):
         # also raises Exception if object array with NA values
         # warning here just in case -- previously __setitem__ was
@@ -3034,6 +3042,7 @@ def _getitem_bool_array(self, key):
         indexer = key.nonzero()[0]
         return self._take(indexer, axis=0)
 
+    # indexing method
     def _getitem_multilevel(self, key):
         loc = self.columns.get_loc(key)
         if isinstance(loc, (slice, Series, np.ndarray, Index)):
@@ -3071,6 +3080,7 @@ def _getitem_multilevel(self, key):
         else:
             return self._get_item_cache(key)
 
+    # indexing method
     def _getitem_frame(self, key):
         if key.values.size and not is_bool_dtype(key.values):
             raise ValueError("Must pass DataFrame with boolean values only")
@@ -3465,6 +3475,7 @@ def _box_col_values(self, values, items):
         klass = self._constructor_sliced
         return klass(values, index=self.index, name=items, fastpath=True)
 
+    # indexing method
     def __setitem__(self, key, value):
         key = com.apply_if_callable(key, self)
 
@@ -3481,10 +3492,12 @@ def __setitem__(self, key, value):
             # set column
             self._set_item(key, value)
 
+    # indexing method
     def _setitem_slice(self, key, value):
         self._check_setitem_copy()
         self.loc._setitem_with_indexer(key, value)
 
+    # indexing method
     def _setitem_array(self, key, value):
         # also raises Exception if object array with NA values
         if com.is_bool_indexer(key):
@@ -3507,6 +3520,7 @@ def _setitem_array(self, key, value):
                 self._check_setitem_copy()
                 self.loc._setitem_with_indexer((slice(None), indexer), value)
 
+    # indexing method
     def _setitem_frame(self, key, value):
         # support boolean setting with DataFrame input, e.g.
         # df[df > df2] = 0
@@ -3544,6 +3558,7 @@ def _ensure_valid_index(self, value):
                 value.index.copy(), axis=1, fill_value=np.nan
             )
 
+    # indexing method
     def _set_item(self, key, value):
         """
         Add series to DataFrame in specified column.
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 4e05dfca43e78..24d750b9a9b71 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -3265,6 +3265,7 @@ def _create_indexer(cls, name, indexer):
             _indexer = functools.partial(indexer, name)
             setattr(cls, name, property(_indexer, doc=indexer.__doc__))
 
+    # indexing method
     def get(self, key, default=None):
         """
         Get item from object for given key (ex: DataFrame column).
@@ -3284,9 +3285,11 @@ def get(self, key, default=None):
         except (KeyError, ValueError, IndexError):
             return default
 
+    # indexing method
     def __getitem__(self, item):
         return self._get_item_cache(item)
 
+    # indexing method
     def _get_item_cache(self, item):
         """Return the cached item, item represents a label indexer."""
         cache = self._item_cache
@@ -3312,6 +3315,7 @@ def _reset_cacher(self):
         if hasattr(self, "_cacher"):
             del self._cacher
 
+    # indexing method
     def _iget_item_cache(self, item):
         """Return the cached item, item represents a positional indexer."""
         ax = self._info_axis
@@ -3386,6 +3390,7 @@ def _clear_item_cache(self, i=None):
         else:
             self._item_cache.clear()
 
+    # indexing method
     def _slice(self, slobj, axis=0, kind=None):
         """
         Construct a slice of this container.
@@ -3402,6 +3407,7 @@ def _slice(self, slobj, axis=0, kind=None):
         result._set_is_copy(self, copy=is_copy)
         return result
 
+    # indexing method
     def _set_item(self, key, value):
         self._data.set(key, value)
         self._clear_item_cache()
@@ -3512,6 +3518,7 @@ def _check_setitem_copy(self, stacklevel=4, t="setting", force=False):
         elif value == "warn":
             warnings.warn(t, com.SettingWithCopyWarning, stacklevel=stacklevel)
 
+    # indexing method
     def __delitem__(self, key):
         """
         Delete item
@@ -3546,6 +3553,7 @@ def __delitem__(self, key):
         except KeyError:
             pass
 
+    # indexing method
     def _take(self, indices, axis=0, is_copy=True):
         """
         Return the elements in the given *positional* indices along an axis.
@@ -3592,6 +3600,7 @@ def _take(self, indices, axis=0, is_copy=True):
 
         return result
 
+    # indexing method
     def take(self, indices, axis=0, is_copy=True, **kwargs):
         """
         Return the elements in the given *positional* indices along an axis.
@@ -3670,6 +3679,7 @@ class  max_speed
         nv.validate_take(tuple(), kwargs)
         return self._take(indices, axis=axis, is_copy=is_copy)
 
+    # indexing method
     def xs(self, key, axis=0, level=None, drop_level=True):
         """
         Return cross-section from the Series/DataFrame.
@@ -5288,9 +5298,6 @@ def _dir_additions(self):
         }
         return super()._dir_additions().union(additions)
 
-    # ----------------------------------------------------------------------
-    # Getting and setting elements
-
     # ----------------------------------------------------------------------
     # Consolidation of internals
 
@@ -9031,6 +9038,7 @@ def _align_series(
 
         return left.__finalize__(self), right.__finalize__(other)
 
+    # indexing method
     def _where(
         self,
         cond,
@@ -10348,6 +10356,7 @@ def describe_1d(data):
         d.columns = data.columns.copy()
         return d
 
+    # TODO: doesn't need to be a method
     def _check_percentile(self, q):
         """
         Validate percentiles (used by describe and quantile).
diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index c31d6538ad2c3..539c884ab05da 100755
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -826,9 +826,6 @@ def _getitem_tuple(self, tup):
         # no shortcut needed
         retval = self.obj
         for i, key in enumerate(tup):
-            if i >= self.obj.ndim:
-                raise IndexingError("Too many indexers")
-
             if com.is_null_slice(key):
                 continue
 
@@ -2114,9 +2111,6 @@ def _getitem_tuple(self, tup):
         retval = self.obj
         axis = 0
         for i, key in enumerate(tup):
-            if i >= self.obj.ndim:
-                raise IndexingError("Too many indexers")
-
             if com.is_null_slice(key):
                 axis += 1
                 continue
diff --git a/pandas/core/series.py b/pandas/core/series.py
index 4b78907e66106..a76639bdddc19 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -991,6 +991,7 @@ def imag(self, v):
 
     # ----------------------------------------------------------------------
 
+    # TODO: can we get rid of this?
     def _unpickle_series_compat(self, state):
         if isinstance(state, dict):
             self._data = state["_data"]
@@ -1028,6 +1029,7 @@ def axes(self):
         """
         return [self.index]
 
+    # indexing method
     def _ixs(self, i, axis=0):
         """
         Return the i-th value or values in the Series by location.
@@ -1065,10 +1067,12 @@ def _ixs(self, i, axis=0):
     def _is_mixed_type(self):
         return False
 
+    # indexing method
     def _slice(self, slobj, axis=0, kind=None):
         slobj = self.index._convert_slice_indexer(slobj, kind=kind or "getitem")
         return self._get_values(slobj)
 
+    # indexing method
     def __getitem__(self, key):
         key = com.apply_if_callable(key, self)
         try:
@@ -1116,6 +1120,7 @@ def __getitem__(self, key):
 
         return self._get_with(key)
 
+    # indexing method
     def _get_with(self, key):
         # other: fancy integer or otherwise
         if isinstance(key, slice):
@@ -1167,6 +1172,7 @@ def _get_with(self, key):
                 return self._get_values(key)
             raise
 
+    # indexing method
     def _get_values_tuple(self, key):
         # mpl hackaround
         if com._any_none(*key):
@@ -1181,6 +1187,7 @@ def _get_values_tuple(self, key):
             self
         )
 
+    # indexing method
     def _get_values(self, indexer):
         try:
             return self._constructor(
@@ -1189,6 +1196,7 @@ def _get_values(self, indexer):
         except Exception:
             return self._values[indexer]
 
+    # indexing method
     def __setitem__(self, key, value):
         key = com.apply_if_callable(key, self)
 
@@ -1247,6 +1255,7 @@ def setitem(key, value):
         if cacher_needs_updating:
             self._maybe_update_cacher()
 
+    # indexing method
     def _set_with_engine(self, key, value):
         values = self._values
         try:
@@ -1256,6 +1265,7 @@ def _set_with_engine(self, key, value):
             values[self.index.get_loc(key)] = value
             return
 
+    # indexing method
     def _set_with(self, key, value):
         # other: fancy integer or otherwise
         if isinstance(key, slice):
@@ -1291,6 +1301,7 @@ def _set_with(self, key, value):
             else:
                 self._set_labels(key, value)
 
+    # indexing method
     def _set_labels(self, key, value):
         if isinstance(key, Index):
             key = key.values
@@ -1302,6 +1313,7 @@ def _set_labels(self, key, value):
             raise ValueError("%s not contained in the index" % str(key[mask]))
         self._set_values(indexer, value)
 
+    # indexing method
     def _set_values(self, key, value):
         if isinstance(key, Series):
             key = key._values
@@ -1365,6 +1377,7 @@ def repeat(self, repeats, axis=None):
         new_values = self._values.repeat(repeats)
         return self._constructor(new_values, index=new_index).__finalize__(self)
 
+    # indexing method
     def get_value(self, label, takeable=False):
         """
         Quickly retrieve single value at passed index label.
@@ -1390,6 +1403,7 @@ def get_value(self, label, takeable=False):
         )
         return self._get_value(label, takeable=takeable)
 
+    # indexing method
     def _get_value(self, label, takeable=False):
         if takeable is True:
             return com.maybe_box_datetimelike(self._values[label])
@@ -1397,6 +1411,7 @@ def _get_value(self, label, takeable=False):
 
     _get_value.__doc__ = get_value.__doc__
 
+    # indexing method
     def set_value(self, label, value, takeable=False):
         """
         Quickly set single value at passed label.
@@ -1430,6 +1445,7 @@ def set_value(self, label, value, takeable=False):
         )
         return self._set_value(label, value, takeable=takeable)
 
+    # indexing method
     def _set_value(self, label, value, takeable=False):
         try:
             if takeable:
@@ -4364,6 +4380,7 @@ def memory_usage(self, index=True, deep=False):
             v += self.index.memory_usage(deep=deep)
         return v
 
+    # indexing method
     @Appender(generic.NDFrame._take.__doc__)
     def _take(self, indices, axis=0, is_copy=False):
 

From 31f44f5bde82b38fce0cdf213e0047bfb3ee7b1e Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 10:58:49 -0700
Subject: [PATCH 02/12] types

---
 pandas/core/indexing.py | 34 ++++++++++------------------------
 1 file changed, 10 insertions(+), 24 deletions(-)

diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index 4650ded627d63..89f02df19e93f 100755
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -143,10 +143,7 @@ def __getitem__(self, key):
             key = com.apply_if_callable(key, self.obj)
             return self._getitem_axis(key, axis=axis)
 
-    def _get_label(self, label, axis=None):
-        if axis is None:
-            axis = self.axis or 0
-
+    def _get_label(self, label, axis: int):
         if self.ndim == 1:
             # for perf reasons we want to try _xs first
             # as its basically direct indexing
@@ -161,9 +158,7 @@ def _get_label(self, label, axis=None):
     def _get_loc(self, key, axis: int):
         return self.obj._ixs(key, axis=axis)
 
-    def _slice(self, obj, axis=None, kind=None):
-        if axis is None:
-            axis = self.axis
+    def _slice(self, obj, axis: int, kind=None):
         return self.obj._slice(obj, axis=axis, kind=kind)
 
     def _get_setitem_indexer(self, key):
@@ -879,10 +874,10 @@ def _convert_for_reindex(self, key, axis: int):
 
     def _handle_lowerdim_multi_index_axis0(self, tup):
         # we have an axis0 multi-index, handle or raise
-
+        axis = self.axis or 0
         try:
             # fast path for series or for tup devoid of slices
-            return self._get_label(tup, axis=self.axis)
+            return self._get_label(tup, axis=axis)
         except TypeError:
             # slices are unhashable
             pass
@@ -980,7 +975,8 @@ def _getitem_nested_tuple(self, tup):
 
             # this is a series with a multi-index specified a tuple of
             # selectors
-            return self._getitem_axis(tup, axis=self.axis)
+            axis = self.axis or 0
+            return self._getitem_axis(tup, axis=axis)
 
         # handle the multi-axis by taking sections and reducing
         # this is iterative
@@ -1007,11 +1003,7 @@ def _getitem_nested_tuple(self, tup):
 
         return obj
 
-    def _getitem_axis(self, key, axis=None):
-
-        if axis is None:
-            axis = self.axis or 0
-
+    def _getitem_axis(self, key, axis: int):
         if is_iterator(key):
             key = list(key)
         self._validate_key(key, axis)
@@ -1436,7 +1428,7 @@ def _is_scalar_access(self, key):
     def _getitem_scalar(self, key):
         raise NotImplementedError()
 
-    def _getitem_axis(self, key, axis=None):
+    def _getitem_axis(self, key, axis: int):
         raise NotImplementedError()
 
     def _getbool_axis(self, key, axis: int):
@@ -1783,10 +1775,7 @@ def _get_partial_string_timestamp_match_key(self, key, labels):
 
         return key
 
-    def _getitem_axis(self, key, axis=None):
-        if axis is None:
-            axis = self.axis or 0
-
+    def _getitem_axis(self, key, axis: int):
         key = item_from_zerodim(key)
         if is_iterator(key):
             key = list(key)
@@ -2137,10 +2126,7 @@ def _get_list_axis(self, key, axis: int):
             # re-raise with different error message
             raise IndexError("positional indexers are out-of-bounds")
 
-    def _getitem_axis(self, key, axis=None):
-        if axis is None:
-            axis = self.axis or 0
-
+    def _getitem_axis(self, key, axis: int):
         if isinstance(key, slice):
             return self._get_slice_axis(key, axis=axis)
 

From 6148c028790616715074a90a9c0a0a66e8ff4165 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 11:58:48 -0700
Subject: [PATCH 03/12] change hasattr check

---
 pandas/core/generic.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index c23e7a02b0ea0..6d8d84bf8af02 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -3502,7 +3502,7 @@ def __delitem__(self, key):
         deleted = False
 
         maybe_shortcut = False
-        if hasattr(self, "columns") and isinstance(self.columns, MultiIndex):
+        if self.ndim == 2 and isinstance(self.columns, MultiIndex):
             try:
                 maybe_shortcut = key not in self.columns._engine
             except TypeError:

From 152d884b48a36cfb864481235f83d11510696ad1 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 12:08:51 -0700
Subject: [PATCH 04/12] refactor out setitem_with_indexer_missing

---
 pandas/core/indexing.py | 119 ++++++++++++++++++++--------------------
 1 file changed, 60 insertions(+), 59 deletions(-)

diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index 89f02df19e93f..5fdfea5366243 100755
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -401,65 +401,7 @@ def _setitem_with_indexer(self, indexer, value):
             indexer, missing = convert_missing_indexer(indexer)
 
             if missing:
-
-                # reindex the axis to the new value
-                # and set inplace
-                if self.ndim == 1:
-                    index = self.obj.index
-                    new_index = index.insert(len(index), indexer)
-
-                    # we have a coerced indexer, e.g. a float
-                    # that matches in an Int64Index, so
-                    # we will not create a duplicate index, rather
-                    # index to that element
-                    # e.g. 0.0 -> 0
-                    # GH12246
-                    if index.is_unique:
-                        new_indexer = index.get_indexer([new_index[-1]])
-                        if (new_indexer != -1).any():
-                            return self._setitem_with_indexer(new_indexer, value)
-
-                    # this preserves dtype of the value
-                    new_values = Series([value])._values
-                    if len(self.obj._values):
-                        # GH#22717 handle casting compatibility that np.concatenate
-                        #  does incorrectly
-                        new_values = _concat_compat([self.obj._values, new_values])
-                    self.obj._data = self.obj._constructor(
-                        new_values, index=new_index, name=self.obj.name
-                    )._data
-                    self.obj._maybe_update_cacher(clear=True)
-                    return self.obj
-
-                elif self.ndim == 2:
-
-                    # no columns and scalar
-                    if not len(self.obj.columns):
-                        raise ValueError(
-                            "cannot set a frame with no defined " "columns"
-                        )
-
-                    # append a Series
-                    if isinstance(value, Series):
-
-                        value = value.reindex(index=self.obj.columns, copy=True)
-                        value.name = indexer
-
-                    # a list-list
-                    else:
-
-                        # must have conforming columns
-                        if is_list_like_indexer(value):
-                            if len(value) != len(self.obj.columns):
-                                raise ValueError(
-                                    "cannot set a row with " "mismatched columns"
-                                )
-
-                        value = Series(value, index=self.obj.columns, name=indexer)
-
-                    self.obj._data = self.obj.append(value)._data
-                    self.obj._maybe_update_cacher(clear=True)
-                    return self.obj
+                return self._setitem_with_indexer_missing(indexer, value)
 
         # set
         item_labels = self.obj._get_axis(info_axis)
@@ -651,6 +593,65 @@ def setter(item, v):
             self.obj._data = self.obj._data.setitem(indexer=indexer, value=value)
             self.obj._maybe_update_cacher(clear=True)
 
+    def _setitem_with_indexer_missing(self, indexer, value):
+        """
+        Insert new row(s) or column(s) into the Series or DataFrame.
+        """
+        from pandas import Series
+
+        # reindex the axis to the new value
+        # and set inplace
+        if self.ndim == 1:
+            index = self.obj.index
+            new_index = index.insert(len(index), indexer)
+
+            # we have a coerced indexer, e.g. a float
+            # that matches in an Int64Index, so
+            # we will not create a duplicate index, rather
+            # index to that element
+            # e.g. 0.0 -> 0
+            # GH#12246
+            if index.is_unique:
+                new_indexer = index.get_indexer([new_index[-1]])
+                if (new_indexer != -1).any():
+                    return self._setitem_with_indexer(new_indexer, value)
+
+            # this preserves dtype of the value
+            new_values = Series([value])._values
+            if len(self.obj._values):
+                # GH#22717 handle casting compatibility that np.concatenate
+                #  does incorrectly
+                new_values = _concat_compat([self.obj._values, new_values])
+            self.obj._data = self.obj._constructor(
+                new_values, index=new_index, name=self.obj.name
+            )._data
+            self.obj._maybe_update_cacher(clear=True)
+            return self.obj
+
+        elif self.ndim == 2:
+
+            if not len(self.obj.columns):
+                # no columns and scalar
+                raise ValueError("cannot set a frame with no defined columns")
+
+            if isinstance(value, ABCSeries):
+                # append a Series
+                value = value.reindex(index=self.obj.columns, copy=True)
+                value.name = indexer
+
+            else:
+                # a list-list
+                if is_list_like_indexer(value):
+                    # must have conforming columns
+                    if len(value) != len(self.obj.columns):
+                        raise ValueError("cannot set a row with mismatched columns")
+
+                value = Series(value, index=self.obj.columns, name=indexer)
+
+            self.obj._data = self.obj.append(value)._data
+            self.obj._maybe_update_cacher(clear=True)
+            return self.obj
+
     def _align_series(self, indexer, ser, multiindex_indexer=False):
         """
         Parameters

From 35a76c8dfe369df585b608c09ef564e0dff8dd93 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 13:49:50 -0700
Subject: [PATCH 05/12] remove no-longer-necessary Panel-compat code from
 GH#10360

---
 pandas/core/indexing.py | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index 5fdfea5366243..3c1ec0eb6ac56 100755
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -325,19 +325,6 @@ def _setitem_with_indexer(self, indexer, value):
                 val = list(value.values()) if isinstance(value, dict) else value
                 take_split_path = not blk._can_hold_element(val)
 
-        if isinstance(indexer, tuple) and len(indexer) == len(self.obj.axes):
-
-            for i, ax in zip(indexer, self.obj.axes):
-
-                # if we have any multi-indexes that have non-trivial slices
-                # (not null slices) then we must take the split path, xref
-                # GH 10360
-                if isinstance(ax, MultiIndex) and not (
-                    is_integer(i) or com.is_null_slice(i)
-                ):
-                    take_split_path = True
-                    break
-
         if isinstance(indexer, tuple):
             nindexer = []
             for i, idx in enumerate(indexer):

From 44eb31780354add43451f0d729d619309c8e76a0 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 15:11:27 -0700
Subject: [PATCH 06/12] cleanups becasue we know 2d

---
 pandas/core/indexing.py | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index 3c1ec0eb6ac56..93c5fc85ba9cd 100755
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -395,6 +395,9 @@ def _setitem_with_indexer(self, indexer, value):
 
         # align and set the values
         if take_split_path:
+            # Above we only set take_split_path to True for 2D cases
+            assert self.ndim == 2
+            assert info_axis == 1
 
             if not isinstance(indexer, tuple):
                 indexer = self._tuplify(indexer)
@@ -448,11 +451,8 @@ def _setitem_with_indexer(self, indexer, value):
             # non-mi
             else:
                 plane_indexer = indexer[:info_axis] + indexer[info_axis + 1 :]
-                if info_axis > 0:
-                    plane_axis = self.obj.axes[:info_axis][0]
-                    lplane_indexer = length_of_indexer(plane_indexer[0], plane_axis)
-                else:
-                    lplane_indexer = 0
+                plane_axis = self.obj.axes[:info_axis][0]
+                lplane_indexer = length_of_indexer(plane_indexer[0], plane_axis)
 
             def setter(item, v):
                 s = self.obj[item]
@@ -502,9 +502,7 @@ def setter(item, v):
                 # hasattr first, to avoid coercing to ndarray without reason.
                 # But we may be relying on the ndarray coercion to check ndim.
                 # Why not just convert to an ndarray earlier on if needed?
-                elif (hasattr(value, "ndim") and value.ndim == 2) or (
-                    not hasattr(value, "ndim") and np.array(value).ndim
-                ) == 2:
+                elif np.ndim(value) == 2:
 
                     # note that this coerces the dtype if we are mixed
                     # GH 7551

From bf170ebc107ca05a1c611b48e76389826a85f555 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 16:31:51 -0700
Subject: [PATCH 07/12] all extant usages of _get_loc pass an int key

---
 pandas/core/indexing.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/core/indexing.py b/pandas/core/indexing.py
index 93c5fc85ba9cd..431b25ab2d107 100755
--- a/pandas/core/indexing.py
+++ b/pandas/core/indexing.py
@@ -155,7 +155,7 @@ def _get_label(self, label, axis: int):
 
         return self.obj._xs(label, axis=axis)
 
-    def _get_loc(self, key, axis: int):
+    def _get_loc(self, key: int, axis: int):
         return self.obj._ixs(key, axis=axis)
 
     def _slice(self, obj, axis: int, kind=None):

From 4e6e1135f8e555df36f79eb977d20423c40a876e Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 16:35:01 -0700
Subject: [PATCH 08/12] trim never-reached branches

---
 pandas/core/frame.py | 76 +++++++++++++++++---------------------------
 1 file changed, 29 insertions(+), 47 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 6035f5c4461aa..2af9c72c54058 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2895,11 +2895,11 @@ def _set_value(self, index, col, value, takeable=False):
     _set_value.__doc__ = set_value.__doc__
 
     # indexing method
-    def _ixs(self, i, axis=0):
+    def _ixs(self, i: int, axis: int = 0):
         """
         Parameters
         ----------
-        i : int, slice, or sequence of integers
+        i : int
         axis : int
 
         Notes
@@ -2908,59 +2908,41 @@ def _ixs(self, i, axis=0):
         """
         # irow
         if axis == 0:
-            if isinstance(i, slice):
-                return self[i]
-            else:
-                label = self.index[i]
-                if isinstance(label, Index):
-                    # a location index by definition
-                    result = self.take(i, axis=axis)
-                    copy = True
-                else:
-                    new_values = self._data.fast_xs(i)
-                    if is_scalar(new_values):
-                        return new_values
-
-                    # if we are a copy, mark as such
-                    copy = (
-                        isinstance(new_values, np.ndarray) and new_values.base is None
-                    )
-                    result = self._constructor_sliced(
-                        new_values,
-                        index=self.columns,
-                        name=self.index[i],
-                        dtype=new_values.dtype,
-                    )
-                result._set_is_copy(self, copy=copy)
-                return result
+            label = self.index[i]
+            new_values = self._data.fast_xs(i)
+            if is_scalar(new_values):
+                return new_values
+
+            # if we are a copy, mark as such
+            copy = isinstance(new_values, np.ndarray) and new_values.base is None
+            result = self._constructor_sliced(
+                new_values,
+                index=self.columns,
+                name=self.index[i],
+                dtype=new_values.dtype,
+            )
+            result._set_is_copy(self, copy=copy)
+            return result
 
         # icol
         else:
             label = self.columns[i]
-            if isinstance(i, slice):
-                # need to return view
-                lab_slice = slice(label[0], label[-1])
-                return self.loc[:, lab_slice]
-            else:
-                if isinstance(label, Index):
-                    return self._take(i, axis=1)
+            index_len = len(self.index)
 
-                index_len = len(self.index)
+            # if the values returned are not the same length
+            # 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)
 
-                # if the values returned are not the same length
-                # 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)
+            if index_len and not len(values):
+                values = np.array([np.nan] * index_len, dtype=object)
+            result = self._box_col_values(values, label)
 
-                if index_len and not len(values):
-                    values = np.array([np.nan] * index_len, dtype=object)
-                result = self._box_col_values(values, label)
+            # this is a cached value, mark it so
+            result._set_as_cached(label, self)
 
-                # this is a cached value, mark it so
-                result._set_as_cached(label, self)
-
-                return result
+            return result
 
     # indexing method
     def __getitem__(self, key):

From df7cac46a90ed23da88a2441d4bdb040b5b26048 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 16:44:54 -0700
Subject: [PATCH 09/12] remove marks

---
 pandas/core/frame.py   | 14 --------------
 pandas/core/generic.py | 11 -----------
 pandas/core/series.py  | 16 ----------------
 3 files changed, 41 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 2af9c72c54058..0b04362cbba2d 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2784,7 +2784,6 @@ def _unpickle_matrix_compat(self, state):  # pragma: no cover
     # ----------------------------------------------------------------------
     # Getting and setting elements
 
-    # indexing method
     def get_value(self, index, col, takeable=False):
         """
         Quickly retrieve single value at passed column and index.
@@ -2812,7 +2811,6 @@ def get_value(self, index, col, takeable=False):
         )
         return self._get_value(index, col, takeable=takeable)
 
-    # indexing method
     def _get_value(self, index, col, takeable=False):
 
         if takeable:
@@ -2840,7 +2838,6 @@ def _get_value(self, index, col, takeable=False):
 
     _get_value.__doc__ = get_value.__doc__
 
-    # indexing method
     def set_value(self, index, col, value, takeable=False):
         """
         Put single value at passed column and index.
@@ -2870,7 +2867,6 @@ def set_value(self, index, col, value, takeable=False):
         )
         return self._set_value(index, col, value, takeable=takeable)
 
-    # indexing method
     def _set_value(self, index, col, value, takeable=False):
         try:
             if takeable is True:
@@ -2894,7 +2890,6 @@ def _set_value(self, index, col, value, takeable=False):
 
     _set_value.__doc__ = set_value.__doc__
 
-    # indexing method
     def _ixs(self, i: int, axis: int = 0):
         """
         Parameters
@@ -2944,7 +2939,6 @@ def _ixs(self, i: int, axis: int = 0):
 
             return result
 
-    # indexing method
     def __getitem__(self, key):
         key = lib.item_from_zerodim(key)
         key = com.apply_if_callable(key, self)
@@ -3000,7 +2994,6 @@ def __getitem__(self, key):
 
         return data
 
-    # indexing method
     def _getitem_bool_array(self, key):
         # also raises Exception if object array with NA values
         # warning here just in case -- previously __setitem__ was
@@ -3024,7 +3017,6 @@ def _getitem_bool_array(self, key):
         indexer = key.nonzero()[0]
         return self._take(indexer, axis=0)
 
-    # indexing method
     def _getitem_multilevel(self, key):
         loc = self.columns.get_loc(key)
         if isinstance(loc, (slice, Series, np.ndarray, Index)):
@@ -3062,7 +3054,6 @@ def _getitem_multilevel(self, key):
         else:
             return self._get_item_cache(key)
 
-    # indexing method
     def _getitem_frame(self, key):
         if key.values.size and not is_bool_dtype(key.values):
             raise ValueError("Must pass DataFrame with boolean values only")
@@ -3457,7 +3448,6 @@ def _box_col_values(self, values, items):
         klass = self._constructor_sliced
         return klass(values, index=self.index, name=items, fastpath=True)
 
-    # indexing method
     def __setitem__(self, key, value):
         key = com.apply_if_callable(key, self)
 
@@ -3474,12 +3464,10 @@ def __setitem__(self, key, value):
             # set column
             self._set_item(key, value)
 
-    # indexing method
     def _setitem_slice(self, key, value):
         self._check_setitem_copy()
         self.loc._setitem_with_indexer(key, value)
 
-    # indexing method
     def _setitem_array(self, key, value):
         # also raises Exception if object array with NA values
         if com.is_bool_indexer(key):
@@ -3502,7 +3490,6 @@ def _setitem_array(self, key, value):
                 self._check_setitem_copy()
                 self.loc._setitem_with_indexer((slice(None), indexer), value)
 
-    # indexing method
     def _setitem_frame(self, key, value):
         # support boolean setting with DataFrame input, e.g.
         # df[df > df2] = 0
@@ -3540,7 +3527,6 @@ def _ensure_valid_index(self, value):
                 value.index.copy(), axis=1, fill_value=np.nan
             )
 
-    # indexing method
     def _set_item(self, key, value):
         """
         Add series to DataFrame in specified column.
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 6d8d84bf8af02..ae42adbf10f5b 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -3241,7 +3241,6 @@ def _create_indexer(cls, name, indexer):
             _indexer = functools.partial(indexer, name)
             setattr(cls, name, property(_indexer, doc=indexer.__doc__))
 
-    # indexing method
     def get(self, key, default=None):
         """
         Get item from object for given key (ex: DataFrame column).
@@ -3261,11 +3260,9 @@ def get(self, key, default=None):
         except (KeyError, ValueError, IndexError):
             return default
 
-    # indexing method
     def __getitem__(self, item):
         return self._get_item_cache(item)
 
-    # indexing method
     def _get_item_cache(self, item):
         """Return the cached item, item represents a label indexer."""
         cache = self._item_cache
@@ -3291,7 +3288,6 @@ def _reset_cacher(self):
         if hasattr(self, "_cacher"):
             del self._cacher
 
-    # indexing method
     def _iget_item_cache(self, item):
         """Return the cached item, item represents a positional indexer."""
         ax = self._info_axis
@@ -3366,7 +3362,6 @@ def _clear_item_cache(self, i=None):
         else:
             self._item_cache.clear()
 
-    # indexing method
     def _slice(self, slobj, axis=0, kind=None):
         """
         Construct a slice of this container.
@@ -3383,7 +3378,6 @@ def _slice(self, slobj, axis=0, kind=None):
         result._set_is_copy(self, copy=is_copy)
         return result
 
-    # indexing method
     def _set_item(self, key, value):
         self._data.set(key, value)
         self._clear_item_cache()
@@ -3494,7 +3488,6 @@ def _check_setitem_copy(self, stacklevel=4, t="setting", force=False):
         elif value == "warn":
             warnings.warn(t, com.SettingWithCopyWarning, stacklevel=stacklevel)
 
-    # indexing method
     def __delitem__(self, key):
         """
         Delete item
@@ -3529,7 +3522,6 @@ def __delitem__(self, key):
         except KeyError:
             pass
 
-    # indexing method
     def _take(self, indices, axis=0, is_copy=True):
         """
         Return the elements in the given *positional* indices along an axis.
@@ -3576,7 +3568,6 @@ def _take(self, indices, axis=0, is_copy=True):
 
         return result
 
-    # indexing method
     def take(self, indices, axis=0, is_copy=True, **kwargs):
         """
         Return the elements in the given *positional* indices along an axis.
@@ -3655,7 +3646,6 @@ class  max_speed
         nv.validate_take(tuple(), kwargs)
         return self._take(indices, axis=axis, is_copy=is_copy)
 
-    # indexing method
     def xs(self, key, axis=0, level=None, drop_level=True):
         """
         Return cross-section from the Series/DataFrame.
@@ -9014,7 +9004,6 @@ def _align_series(
 
         return left.__finalize__(self), right.__finalize__(other)
 
-    # indexing method
     def _where(
         self,
         cond,
diff --git a/pandas/core/series.py b/pandas/core/series.py
index 4b50e995f5a20..7cb287b0d8375 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -1029,7 +1029,6 @@ def axes(self):
         """
         return [self.index]
 
-    # indexing method
     def _ixs(self, i, axis=0):
         """
         Return the i-th value or values in the Series by location.
@@ -1067,12 +1066,10 @@ def _ixs(self, i, axis=0):
     def _is_mixed_type(self):
         return False
 
-    # indexing method
     def _slice(self, slobj, axis=0, kind=None):
         slobj = self.index._convert_slice_indexer(slobj, kind=kind or "getitem")
         return self._get_values(slobj)
 
-    # indexing method
     def __getitem__(self, key):
         key = com.apply_if_callable(key, self)
         try:
@@ -1120,7 +1117,6 @@ def __getitem__(self, key):
 
         return self._get_with(key)
 
-    # indexing method
     def _get_with(self, key):
         # other: fancy integer or otherwise
         if isinstance(key, slice):
@@ -1172,7 +1168,6 @@ def _get_with(self, key):
                 return self._get_values(key)
             raise
 
-    # indexing method
     def _get_values_tuple(self, key):
         # mpl hackaround
         if com._any_none(*key):
@@ -1187,7 +1182,6 @@ def _get_values_tuple(self, key):
             self
         )
 
-    # indexing method
     def _get_values(self, indexer):
         try:
             return self._constructor(
@@ -1196,7 +1190,6 @@ def _get_values(self, indexer):
         except Exception:
             return self._values[indexer]
 
-    # indexing method
     def __setitem__(self, key, value):
         key = com.apply_if_callable(key, self)
 
@@ -1255,7 +1248,6 @@ def setitem(key, value):
         if cacher_needs_updating:
             self._maybe_update_cacher()
 
-    # indexing method
     def _set_with_engine(self, key, value):
         values = self._values
         try:
@@ -1265,7 +1257,6 @@ def _set_with_engine(self, key, value):
             values[self.index.get_loc(key)] = value
             return
 
-    # indexing method
     def _set_with(self, key, value):
         # other: fancy integer or otherwise
         if isinstance(key, slice):
@@ -1308,7 +1299,6 @@ def _set_with(self, key, value):
             else:
                 self._set_labels(key, value)
 
-    # indexing method
     def _set_labels(self, key, value):
         if isinstance(key, Index):
             key = key.values
@@ -1320,7 +1310,6 @@ def _set_labels(self, key, value):
             raise ValueError("%s not contained in the index" % str(key[mask]))
         self._set_values(indexer, value)
 
-    # indexing method
     def _set_values(self, key, value):
         if isinstance(key, Series):
             key = key._values
@@ -1384,7 +1373,6 @@ def repeat(self, repeats, axis=None):
         new_values = self._values.repeat(repeats)
         return self._constructor(new_values, index=new_index).__finalize__(self)
 
-    # indexing method
     def get_value(self, label, takeable=False):
         """
         Quickly retrieve single value at passed index label.
@@ -1410,7 +1398,6 @@ def get_value(self, label, takeable=False):
         )
         return self._get_value(label, takeable=takeable)
 
-    # indexing method
     def _get_value(self, label, takeable=False):
         if takeable is True:
             return com.maybe_box_datetimelike(self._values[label])
@@ -1418,7 +1405,6 @@ def _get_value(self, label, takeable=False):
 
     _get_value.__doc__ = get_value.__doc__
 
-    # indexing method
     def set_value(self, label, value, takeable=False):
         """
         Quickly set single value at passed label.
@@ -1452,7 +1438,6 @@ def set_value(self, label, value, takeable=False):
         )
         return self._set_value(label, value, takeable=takeable)
 
-    # indexing method
     def _set_value(self, label, value, takeable=False):
         try:
             if takeable:
@@ -4387,7 +4372,6 @@ def memory_usage(self, index=True, deep=False):
             v += self.index.memory_usage(deep=deep)
         return v
 
-    # indexing method
     @Appender(generic.NDFrame._take.__doc__)
     def _take(self, indices, axis=0, is_copy=False):
 

From 5fb03e127bce9a3a8a83ac61c0b210c67b58e74a Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 16:47:19 -0700
Subject: [PATCH 10/12] simplify ixs

---
 pandas/core/series.py | 29 ++++++++---------------------
 1 file changed, 8 insertions(+), 21 deletions(-)

diff --git a/pandas/core/series.py b/pandas/core/series.py
index 7cb287b0d8375..ad5aaa5e96a71 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -1029,38 +1029,25 @@ def axes(self):
         """
         return [self.index]
 
-    def _ixs(self, i, axis=0):
+    def _ixs(self, i: int, axis: int = 0):
         """
         Return the i-th value or values in the Series by location.
 
         Parameters
         ----------
-        i : int, slice, or sequence of integers
+        i : int
 
         Returns
         -------
         scalar (int) or Series (slice, sequence)
         """
-        try:
 
-            # dispatch to the values if we need
-            values = self._values
-            if isinstance(values, np.ndarray):
-                return libindex.get_value_at(values, i)
-            else:
-                return values[i]
-        except IndexError:
-            raise
-        except Exception:
-            if isinstance(i, slice):
-                indexer = self.index._convert_slice_indexer(i, kind="iloc")
-                return self._get_values(indexer)
-            else:
-                label = self.index[i]
-                if isinstance(label, Index):
-                    return self.take(i, axis=axis, convert=True)
-                else:
-                    return libindex.get_value_at(self, i)
+        # dispatch to the values if we need
+        values = self._values
+        if isinstance(values, np.ndarray):
+            return libindex.get_value_at(values, i)
+        else:
+            return values[i]
 
     @property
     def _is_mixed_type(self):

From 75e0bc11b880087c3b082100c390334a7b634ea0 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Thu, 11 Jul 2019 16:48:41 -0700
Subject: [PATCH 11/12] remove comments

---
 pandas/core/frame.py   | 1 -
 pandas/core/generic.py | 1 -
 pandas/core/series.py  | 1 -
 3 files changed, 3 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 0b04362cbba2d..168cd6a8ebb67 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2751,7 +2751,6 @@ def transpose(self, *args, **kwargs):
     # ----------------------------------------------------------------------
     # Picklability
 
-    # TODO: can we get rid of these?
     # legacy pickle formats
     def _unpickle_frame_compat(self, state):  # pragma: no cover
         if len(state) == 2:  # pragma: no cover
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index ae42adbf10f5b..ceef7dd2a5a7d 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -10321,7 +10321,6 @@ def describe_1d(data):
         d.columns = data.columns.copy()
         return d
 
-    # TODO: doesn't need to be a method
     def _check_percentile(self, q):
         """
         Validate percentiles (used by describe and quantile).
diff --git a/pandas/core/series.py b/pandas/core/series.py
index ad5aaa5e96a71..fa9f79d7d2683 100644
--- a/pandas/core/series.py
+++ b/pandas/core/series.py
@@ -991,7 +991,6 @@ def imag(self, v):
 
     # ----------------------------------------------------------------------
 
-    # TODO: can we get rid of this?
     def _unpickle_series_compat(self, state):
         if isinstance(state, dict):
             self._data = state["_data"]

From 437f8fed5a9754b26a5099f04d631c8ef39a0995 Mon Sep 17 00:00:00 2001
From: jbrockmendel <jbrockmendel@gmail.com>
Date: Fri, 12 Jul 2019 07:55:22 -0700
Subject: [PATCH 12/12] dont use index_len

---
 pandas/core/frame.py | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 7f4ecde7557fc..53cb0cedc208b 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -2921,7 +2921,6 @@ def _ixs(self, i: int, axis: int = 0):
         # icol
         else:
             label = self.columns[i]
-            index_len = len(self.index)
 
             # if the values returned are not the same length
             # as the index (iow a not found value), iget returns
@@ -2929,8 +2928,8 @@ def _ixs(self, i: int, axis: int = 0):
             # a numpy error (as numpy should really raise)
             values = self._data.iget(i)
 
-            if index_len and not len(values):
-                values = np.array([np.nan] * index_len, dtype=object)
+            if len(self.index) and not len(values):
+                values = np.array([np.nan] * len(self.index), dtype=object)
             result = self._box_col_values(values, label)
 
             # this is a cached value, mark it so