From 2a72b6000fa915938933afc36fa322988a5f013e Mon Sep 17 00:00:00 2001
From: Tom Neep <tom.neep@cern.ch>
Date: Mon, 26 Feb 2018 16:23:38 +0100
Subject: [PATCH] BUG: fix to_latex() when using MultiIndex with NaN in
 (#14249)

---
 doc/source/whatsnew/v0.23.0.txt          |  1 +
 pandas/io/formats/latex.py               | 14 ++++++++------
 pandas/tests/io/formats/test_to_latex.py | 22 ++++++++++++++++++++++
 3 files changed, 31 insertions(+), 6 deletions(-)

diff --git a/doc/source/whatsnew/v0.23.0.txt b/doc/source/whatsnew/v0.23.0.txt
index c08e22af295f4..cbfc49d7fa3ab 100644
--- a/doc/source/whatsnew/v0.23.0.txt
+++ b/doc/source/whatsnew/v0.23.0.txt
@@ -966,6 +966,7 @@ I/O
 - Bug in :func:`read_csv` where missing values were not being handled properly when ``keep_default_na=False`` with dictionary ``na_values`` (:issue:`19227`)
 - Bug in :func:`read_sas` where a file with 0 variables gave an ``AttributeError`` incorrectly. Now it gives an ``EmptyDataError`` (:issue:`18184`)
 - Bug in :func:`DataFrame.to_latex()` where pairs of braces meant to serve as invisible placeholders were escaped (:issue:`18667`)
+- Bug in :func:`DataFrame.to_latex()` where a ``NaN`` in a ``MultiIndex`` would cause an ``IndexError`` or incorrect output (:issue:`14249`)
 - Bug in :func:`read_json` where large numeric values were causing an ``OverflowError`` (:issue:`18842`)
 - Bug in :func:`DataFrame.to_parquet` where an exception was raised if the write destination is S3 (:issue:`19134`)
 - :class:`Interval` now supported in :func:`DataFrame.to_excel` for all Excel file types (:issue:`19242`)
diff --git a/pandas/io/formats/latex.py b/pandas/io/formats/latex.py
index 67b0a4f0e034e..e85fcf9edaff4 100644
--- a/pandas/io/formats/latex.py
+++ b/pandas/io/formats/latex.py
@@ -72,7 +72,8 @@ def get_col_type(dtype):
             previous_lev3 = None
             for i, lev in enumerate(self.frame.index.levels):
                 lev2 = lev.format()
-                blank = ' ' * len(lev2[0])
+                blank = (' ' * len(lev2[0]) if lev2 else
+                         ' ' * len(self.fmt.na_rep))
                 # display column names in last index-column
                 if cname and i == lastcol:
                     lev3 = [x if x else '{}' for x in self.frame.columns.names]
@@ -82,15 +83,16 @@ def get_col_type(dtype):
                     lev3.append(lev.name)
                 current_idx_val = None
                 for level_idx in self.frame.index.labels[i]:
+                    idx_val = (lev2[level_idx] if level_idx >= 0 else
+                               self.fmt.na_rep)
                     if ((previous_lev3 is None or
-                        previous_lev3[len(lev3)].isspace()) and
-                            lev2[level_idx] == current_idx_val):
+                         previous_lev3[len(lev3)].isspace()) and
+                            idx_val == current_idx_val):
                         # same index as above row and left index was the same
                         lev3.append(blank)
                     else:
-                        # different value than above or left index different
-                        lev3.append(lev2[level_idx])
-                        current_idx_val = lev2[level_idx]
+                        lev3.append(idx_val)
+                        current_idx_val = idx_val
                 strcols.insert(i, lev3)
                 previous_lev3 = lev3
 
diff --git a/pandas/tests/io/formats/test_to_latex.py b/pandas/tests/io/formats/test_to_latex.py
index 5ebf196be094e..a0f1da855a3b3 100644
--- a/pandas/tests/io/formats/test_to_latex.py
+++ b/pandas/tests/io/formats/test_to_latex.py
@@ -621,3 +621,25 @@ def test_to_latex_multiindex_names(self, name0, name1, axes):
 \end{tabular}
 """ % tuple(list(col_names) + [idx_names_row])
         assert observed == expected
+
+    @pytest.mark.parametrize('one_row', [True, False])
+    def test_to_latex_multiindex_nans(self, one_row):
+        # GH 14249
+        df = pd.DataFrame({'a': [None, 1], 'b': [2, 3], 'c': [4, 5]})
+        if one_row:
+            df = df.iloc[[0]]
+        observed = df.set_index(['a', 'b']).to_latex()
+        expected = r"""\begin{tabular}{llr}
+\toprule
+    &   &  c \\
+a & b &    \\
+\midrule
+NaN & 2 &  4 \\
+"""
+        if not one_row:
+            expected += r"""1.0 & 3 &  5 \\
+"""
+        expected += r"""\bottomrule
+\end{tabular}
+"""
+        assert observed == expected