From 1b3f381e2cf5048e1ab7a0fa5d42588107229419 Mon Sep 17 00:00:00 2001
From: Goyo <goyodiaz@gmail.com>
Date: Mon, 7 Oct 2013 21:44:38 +0200
Subject: [PATCH 1/2] BUG: MultiIndex.get_level_values() replaces NA by another
 value (#5074)

---
 doc/source/release.rst     |  1 +
 pandas/core/index.py       |  7 ++++++-
 pandas/tests/test_index.py | 33 +++++++++++++++++++++++++++++++++
 pandas/tseries/index.py    |  4 ++++
 4 files changed, 44 insertions(+), 1 deletion(-)

diff --git a/doc/source/release.rst b/doc/source/release.rst
index 216b7f2ca6e5a..40ad07aea1ecf 100644
--- a/doc/source/release.rst
+++ b/doc/source/release.rst
@@ -569,6 +569,7 @@ Bug Fixes
   - Fixed a bug where default options were being overwritten in the option
     parser cleaning (:issue:`5121`).
   - Treat a list/ndarray identically for ``iloc`` indexing with list-like (:issue:`5006`)
+  - Fix ``MultiIndex.get_level_values()`` with missing values (:issue:`5074`)
 
 pandas 0.12.0
 -------------
diff --git a/pandas/core/index.py b/pandas/core/index.py
index 8e98cc6fb25bb..5cadd6de2ee5f 100644
--- a/pandas/core/index.py
+++ b/pandas/core/index.py
@@ -393,6 +393,10 @@ def values(self):
     def get_values(self):
         return self.values
 
+    def _na_value(self):
+        # The expected NA value to use with this index.
+        return np.nan
+
     @property
     def is_monotonic(self):
         return self._engine.is_monotonic
@@ -2256,7 +2260,8 @@ def get_level_values(self, level):
         num = self._get_level_number(level)
         unique_vals = self.levels[num]  # .values
         labels = self.labels[num]
-        values = unique_vals.take(labels)
+        values = Index(com.take_1d(unique_vals.values, labels,
+                                   fill_value=unique_vals._na_value()))
         values.name = self.names[num]
         return values
 
diff --git a/pandas/tests/test_index.py b/pandas/tests/test_index.py
index 5404b30af8d1c..7e801c0a202db 100644
--- a/pandas/tests/test_index.py
+++ b/pandas/tests/test_index.py
@@ -1445,6 +1445,39 @@ def test_get_level_values(self):
         expected = self.index.get_level_values(0)
         self.assert_(np.array_equal(result, expected))
 
+    def test_get_level_values_na(self):
+        arrays = [['a', 'b', 'b'], [1, np.nan, 2]]
+        index = pd.MultiIndex.from_arrays(arrays)
+        values = index.get_level_values(1)
+        expected = [1, np.nan, 2]
+        assert_array_equal(values.values.astype(float), expected)
+
+        arrays = [['a', 'b', 'b'], [np.nan, np.nan, 2]]
+        index = pd.MultiIndex.from_arrays(arrays)
+        values = index.get_level_values(1)
+        expected = [np.nan, np.nan, 2]
+        assert_array_equal(values.values.astype(float), expected)
+
+        arrays = [[np.nan, np.nan, np.nan], ['a', np.nan, 1]]
+        index = pd.MultiIndex.from_arrays(arrays)
+        values = index.get_level_values(0)
+        expected = [np.nan, np.nan, np.nan]
+        assert_array_equal(values.values.astype(float), expected)
+        values = index.get_level_values(1)
+        expected = ['a', np.nan, 1]
+        assert_array_equal(values.values, expected)
+
+        arrays = [['a', 'b', 'b'], pd.DatetimeIndex([0, 1, pd.NaT])]
+        index = pd.MultiIndex.from_arrays(arrays)
+        values = index.get_level_values(1)
+        expected = pd.DatetimeIndex([0, 1, pd.NaT])
+        assert_array_equal(values.values, expected.values)
+
+        arrays = [[], []]
+        index = pd.MultiIndex.from_arrays(arrays)
+        values = index.get_level_values(0)
+        self.assertEqual(values.shape, (0,))
+
     def test_reorder_levels(self):
         # this blows up
         assertRaisesRegexp(IndexError, '^Too many levels',
diff --git a/pandas/tseries/index.py b/pandas/tseries/index.py
index 281ac0cc8a35a..0b02235010c9c 100644
--- a/pandas/tseries/index.py
+++ b/pandas/tseries/index.py
@@ -495,6 +495,10 @@ def _mpl_repr(self):
         # how to represent ourselves to matplotlib
         return tslib.ints_to_pydatetime(self.asi8, self.tz)
 
+    def _na_value(self):
+        # The expected NA value to use with this index.
+        return tslib.NaT
+
     def __unicode__(self):
         from pandas.core.format import _format_datetime64
         values = self.values

From 573fee62f334045e5d410ce1e4ee035409e7325b Mon Sep 17 00:00:00 2001
From: Goyo <goyodiaz@gmail.com>
Date: Sun, 6 Oct 2013 14:51:27 +0200
Subject: [PATCH 2/2] Make _na_value a class attribute.

---
 pandas/core/index.py    | 7 +++----
 pandas/tseries/index.py | 5 ++---
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/pandas/core/index.py b/pandas/core/index.py
index 5cadd6de2ee5f..465a0439c6eb3 100644
--- a/pandas/core/index.py
+++ b/pandas/core/index.py
@@ -393,9 +393,8 @@ def values(self):
     def get_values(self):
         return self.values
 
-    def _na_value(self):
-        # The expected NA value to use with this index.
-        return np.nan
+    _na_value = np.nan
+    """The expected NA value to use with this index."""
 
     @property
     def is_monotonic(self):
@@ -2261,7 +2260,7 @@ def get_level_values(self, level):
         unique_vals = self.levels[num]  # .values
         labels = self.labels[num]
         values = Index(com.take_1d(unique_vals.values, labels,
-                                   fill_value=unique_vals._na_value()))
+                                   fill_value=unique_vals._na_value))
         values.name = self.names[num]
         return values
 
diff --git a/pandas/tseries/index.py b/pandas/tseries/index.py
index 0b02235010c9c..ce1bea96b9d4c 100644
--- a/pandas/tseries/index.py
+++ b/pandas/tseries/index.py
@@ -495,9 +495,8 @@ def _mpl_repr(self):
         # how to represent ourselves to matplotlib
         return tslib.ints_to_pydatetime(self.asi8, self.tz)
 
-    def _na_value(self):
-        # The expected NA value to use with this index.
-        return tslib.NaT
+    _na_value = tslib.NaT
+    """The expected NA value to use with this index."""
 
     def __unicode__(self):
         from pandas.core.format import _format_datetime64