deprecate and test has_index_names

Author: chris-b1

pandas/core/format.py

@@ -1833,25 +1833,21 @@ def _format_hierarchical_rows(self):
                                                (list, tuple, np.ndarray, Index)):
                 index_labels = self.index_label
 
+            # MultiIndex columns require an extra row
+            # with index names (blank if None) for
+            # unambigous round-trip
             if isinstance(self.columns, MultiIndex):
                 self.rowcounter += 1
 
             # if index labels are not empty go ahead and dump
             if (any(x is not None for x in index_labels)
                     and self.header is not False):
 
-                # if not self.merge_cells or not isinstance(self.columns, MultiIndex):
-                #     self.rowcounter -= 1
-
                 for cidx, name in enumerate(index_labels):
                     yield ExcelCell(self.rowcounter - 1,
                                     cidx,
                                     name,
                                     header_style)
-            # elif isinstance(self.columns, MultiIndex):
-            #     # if multiIndex columns, leave a blank row
-            #     # regardless of labels
-            #     self.rowcounter += 1
 
             if self.merge_cells:
                 # Format hierarchical rows as merged cells.

pandas/io/excel.py

@@ -135,9 +135,9 @@ def read_excel(io, sheetname=0, **kwds):
         data will be read in as floats: Excel stores all numbers as floats
         internally
     has_index_names : boolean, default False
-        True if the cols defined in index_col have an index name and are
-        not in the header. Index name will be placed on a separate line below
-        the header.
+        DEPCRECATED: for version 0.17+ index names will be automatically inferred
+        based on index_col.  To read Excel output from 0.16.2 and prior that
+        had saved index names, use True.
 
     Returns
     -------
@@ -227,7 +227,6 @@ def parse(self, sheetname=0, header=0, skiprows=None, skip_footer=0,
             Row to use for the column labels of the parsed DataFrame
             If a list of integers is passed those row positions will
             be combined into a ``MultiIndex``
-            Row to use for the column labels of the parsed DataFrame
         skiprows : list-like
             Rows to skip at the beginning (0-indexed)
         skip_footer : int, default 0
@@ -261,8 +260,9 @@ def parse(self, sheetname=0, header=0, skiprows=None, skip_footer=0,
             numeric data will be read in as floats: Excel stores all numbers as
             floats internally.
         has_index_names : boolean, default False
-            True if the cols defined in index_col have an index name and are
-            not in the header
+            DEPCRECATED: for version 0.17+ index names will be automatically inferred
+            based on index_col.  To read Excel output from 0.16.2 and prior that
+            had saved index names, use True.
         verbose : boolean, default False
             Set to True to print a single statement when reading each
             excel sheet.
@@ -277,9 +277,11 @@ def parse(self, sheetname=0, header=0, skiprows=None, skip_footer=0,
         if skipfooter is not None:
             skip_footer = skipfooter
 
-        # if has_index_names:
-        #     warning.warn("The has_index_names argument is deprecated; index names"
-        #                  "will be automatically inferred based on index_col.")
+        if has_index_names:
+            warn("\nThe has_index_names argument is deprecated; index names "
+                 "will be automatically inferred based on index_col.\n"
+                 "This argmument is still necessary if reading Excel output "
+                 "from 0.16.2 or prior with index names.", FutureWarning)
 
         return self._parse_excel(sheetname=sheetname, header=header,
                                  skiprows=skiprows,

pandas/io/tests/test_excel.py

@@ -601,6 +601,42 @@ def test_excel_multindex_roundtrip(self):
                                                 header=list(range(c_idx_levels)))
                             tm.assert_frame_equal(df, act, check_names=check_names)
 
+    def test_excel_oldindex_format(self):
+        #GH 4679
+        _skip_if_no_xlrd()
+        data = np.array([['R0C0', 'R0C1', 'R0C2', 'R0C3', 'R0C4'],
+                         ['R1C0', 'R1C1', 'R1C2', 'R1C3', 'R1C4'],
+                         ['R2C0', 'R2C1', 'R2C2', 'R2C3', 'R2C4'],
+                         ['R3C0', 'R3C1', 'R3C2', 'R3C3', 'R3C4'],
+                         ['R4C0', 'R4C1', 'R4C2', 'R4C3', 'R4C4']])
+        columns = ['C_l0_g0', 'C_l0_g1', 'C_l0_g2', 'C_l0_g3', 'C_l0_g4']
+        mi = MultiIndex(levels=[['R_l0_g0', 'R_l0_g1', 'R_l0_g2', 'R_l0_g3', 'R_l0_g4'],
+                                ['R_l1_g0', 'R_l1_g1', 'R_l1_g2', 'R_l1_g3', 'R_l1_g4']],
+                        labels=[[0, 1, 2, 3, 4], [0, 1, 2, 3, 4]],
+                        names=['R0', 'R1'])
+        si = Index(['R_l0_g0', 'R_l0_g1', 'R_l0_g2', 'R_l0_g3', 'R_l0_g4'],  name='R0')
+
+        in_file = os.path.join(self.dirpath, 'test_index_name_pre17.xlsx')
+
+        expected = pd.DataFrame(data, index=si, columns=columns)
+        with tm.assert_produces_warning(FutureWarning):
+            actual = pd.read_excel(in_file, 'single_names', has_index_names=True)
+        tm.assert_frame_equal(actual, expected)
+
+        expected.index.name = None
+        actual = pd.read_excel(in_file, 'single_no_names')
+        tm.assert_frame_equal(actual, expected)
+
+        expected.index = mi
+        with tm.assert_produces_warning(FutureWarning):
+            actual = pd.read_excel(in_file, 'multi_names', has_index_names=True)
+        tm.assert_frame_equal(actual, expected)
+
+        expected.index.names = [None, None]
+        actual = pd.read_excel(in_file, 'multi_no_names', index_col=[0,1])
+        tm.assert_frame_equal(actual, expected, check_names=False)
+
+
 
 class XlsReaderTests(XlrdTests, tm.TestCase):
     ext = '.xls'
@@ -866,7 +902,6 @@ def test_roundtrip_indexlabels(self):
             reader = ExcelFile(path)
             recons = reader.parse('test1',
                                   index_col=0,
-                                  has_index_names=self.merge_cells
                                   ).astype(np.int64)
             frame.index.names = ['test']
             self.assertEqual(frame.index.names, recons.index.names)
@@ -879,7 +914,6 @@ def test_roundtrip_indexlabels(self):
             reader = ExcelFile(path)
             recons = reader.parse('test1',
                                   index_col=0,
-                                  has_index_names=self.merge_cells
                                   ).astype(np.int64)
             frame.index.names = ['test']
             self.assertEqual(frame.index.names, recons.index.names)
@@ -892,7 +926,6 @@ def test_roundtrip_indexlabels(self):
             reader = ExcelFile(path)
             recons = reader.parse('test1',
                                   index_col=0,
-                                  has_index_names=self.merge_cells
                                   ).astype(np.int64)
             frame.index.names = ['test']
             tm.assert_frame_equal(frame, recons.astype(bool))
@@ -922,8 +955,7 @@ def test_excel_roundtrip_indexname(self):
 
             xf = ExcelFile(path)
             result = xf.parse(xf.sheet_names[0],
-                              index_col=0,
-                              has_index_names=self.merge_cells)
+                              index_col=0)
 
             tm.assert_frame_equal(result, df)
             self.assertEqual(result.index.name, 'foo')
@@ -1010,8 +1042,7 @@ def test_to_excel_multiindex(self):
             frame.to_excel(path, 'test1', merge_cells=self.merge_cells)
             reader = ExcelFile(path)
             df = reader.parse('test1', index_col=[0, 1],
-                              parse_dates=False,
-                              has_index_names=self.merge_cells)
+                              parse_dates=False)
             tm.assert_frame_equal(frame, df)
             self.assertEqual(frame.index.names, df.index.names)
 
@@ -1028,8 +1059,7 @@ def test_to_excel_multiindex_dates(self):
             tsframe.to_excel(path, 'test1', merge_cells=self.merge_cells)
             reader = ExcelFile(path)
             recons = reader.parse('test1',
-                                  index_col=[0, 1],
-                                  has_index_names=self.merge_cells)
+                                  index_col=[0, 1])
 
             tm.assert_frame_equal(tsframe, recons)
             self.assertEqual(recons.index.names, ('time', 'foo'))