TST: Tests and fix for unhandled data types in Excel writers.

Added tests and a fix for unhandled numpy data types in the
Excel writers. Issue #3122.

Author: jmcnamara

pandas/io/excel.py

@@ -16,6 +16,7 @@
 from pandas.core import config
 from pandas.core.common import pprint_thing
 import pandas.compat as compat
+import pandas.core.common as com
 from warnings import warn
 
 __all__ = ["read_excel", "ExcelWriter", "ExcelFile"]
@@ -290,20 +291,20 @@ def __enter__(self):
     def __exit__(self, exc_type, exc_value, traceback):
         self.close()
 
-
 def _trim_excel_header(row):
     # trim header row so auto-index inference works
     # xlrd uses '' , openpyxl None
     while len(row) > 0 and (row[0] == '' or row[0] is None):
         row = row[1:]
     return row
 
-
 def _conv_value(val):
-    # convert value for excel dump
-    if isinstance(val, np.int64):
+    # Convert numpy types to Python types for the Excel writers.
+    if com.is_integer(val):
         val = int(val)
-    elif isinstance(val, np.bool8):
+    elif com.is_float(val):
+        val = float(val)
+    elif com.is_bool(val):
         val = bool(val)
     elif isinstance(val, Period):
         val = "%s" % val
@@ -686,8 +687,6 @@ def write_cells(self, cells, sheet_name=None, startrow=0, startcol=0):
         style_dict = {}
 
         for cell in cells:
-            val = _conv_value(cell.val)
-
             num_format_str = None
             if isinstance(cell.val, datetime.datetime):
                 num_format_str = "YYYY-MM-DD HH:MM:SS"
@@ -709,11 +708,11 @@ def write_cells(self, cells, sheet_name=None, startrow=0, startcol=0):
                                 startrow + cell.mergestart,
                                 startcol + cell.col,
                                 startcol + cell.mergeend,
-                                val, style)
+                                cell.val, style)
             else:
                 wks.write(startrow + cell.row,
                           startcol + cell.col,
-                          val, style)
+                          cell.val, style)
 
     def _convert_to_style(self, style_dict, num_format_str=None):
         """

pandas/io/tests/test_excel.py

@@ -221,7 +221,6 @@ def test_excel_table_sheet_by_index(self):
                                   (self.xlsx1, self.csv1)]:
             self.check_excel_table_sheet_by_index(filename, csvfile)
 
-
     def test_excel_table(self):
         _skip_if_no_xlrd()
 
@@ -405,7 +404,6 @@ def test_mixed(self):
             recons = reader.parse('test1', index_col=0)
             tm.assert_frame_equal(self.mixed_frame, recons)
 
-
     def test_tsframe(self):
         _skip_if_no_xlrd()
         ext = self.ext
@@ -419,45 +417,70 @@ def test_tsframe(self):
             recons = reader.parse('test1')
             tm.assert_frame_equal(df, recons)
 
-    def test_int64(self):
+    def test_int_types(self):
         _skip_if_no_xlrd()
         ext = self.ext
-        path = '__tmp_to_excel_from_excel_int64__.' + ext
-
-        with ensure_clean(path) as path:
-            self.frame['A'][:5] = nan
-
-            self.frame.to_excel(path, 'test1')
-            self.frame.to_excel(path, 'test1', cols=['A', 'B'])
-            self.frame.to_excel(path, 'test1', header=False)
-            self.frame.to_excel(path, 'test1', index=False)
+        path = '__tmp_to_excel_from_excel_int_types__.' + ext
 
-            # Test np.int64, values read come back as float
-            frame = DataFrame(np.random.randint(-10, 10, size=(10, 2)), dtype=np.int64)
-            frame.to_excel(path, 'test1')
-            reader = ExcelFile(path)
-            recons = reader.parse('test1').astype(np.int64)
-            tm.assert_frame_equal(frame, recons, check_dtype=False)
+        for np_type in (np.int8, np.int16, np.int32, np.int64):
 
-    def test_bool(self):
+            with ensure_clean(path) as path:
+                self.frame['A'][:5] = nan
+
+                self.frame.to_excel(path, 'test1')
+                self.frame.to_excel(path, 'test1', cols=['A', 'B'])
+                self.frame.to_excel(path, 'test1', header=False)
+                self.frame.to_excel(path, 'test1', index=False)
+
+                # Test np.int values read come back as float.
+                frame = DataFrame(np.random.randint(-10, 10, size=(10, 2)),
+                                  dtype=np_type)
+                frame.to_excel(path, 'test1')
+                reader = ExcelFile(path)
+                recons = reader.parse('test1').astype(np_type)
+                tm.assert_frame_equal(frame, recons, check_dtype=False)
+
+    def test_float_types(self):
         _skip_if_no_xlrd()
         ext = self.ext
-        path = '__tmp_to_excel_from_excel_bool__.' + ext
+        path = '__tmp_to_excel_from_excel_float_types__.' + ext
 
-        with ensure_clean(path) as path:
-            self.frame['A'][:5] = nan
+        for np_type in (np.float16, np.float32, np.float64):
+            with ensure_clean(path) as path:
+                self.frame['A'][:5] = nan
 
-            self.frame.to_excel(path, 'test1')
-            self.frame.to_excel(path, 'test1', cols=['A', 'B'])
-            self.frame.to_excel(path, 'test1', header=False)
-            self.frame.to_excel(path, 'test1', index=False)
+                self.frame.to_excel(path, 'test1')
+                self.frame.to_excel(path, 'test1', cols=['A', 'B'])
+                self.frame.to_excel(path, 'test1', header=False)
+                self.frame.to_excel(path, 'test1', index=False)
 
-            # Test reading/writing np.bool8, roundtrip only works for xlsx
-            frame = (DataFrame(np.random.randn(10, 2)) >= 0)
-            frame.to_excel(path, 'test1')
-            reader = ExcelFile(path)
-            recons = reader.parse('test1').astype(np.bool8)
-            tm.assert_frame_equal(frame, recons)
+                # Test np.float values read come back as float.
+                frame = DataFrame(np.random.random_sample(10), dtype=np_type)
+                frame.to_excel(path, 'test1')
+                reader = ExcelFile(path)
+                recons = reader.parse('test1').astype(np_type)
+                tm.assert_frame_equal(frame, recons, check_dtype=False)
+
+    def test_bool_types(self):
+        _skip_if_no_xlrd()
+        ext = self.ext
+        path = '__tmp_to_excel_from_excel_bool_types__.' + ext
+
+        for np_type in (np.bool8, np.bool_):
+            with ensure_clean(path) as path:
+                self.frame['A'][:5] = nan
+
+                self.frame.to_excel(path, 'test1')
+                self.frame.to_excel(path, 'test1', cols=['A', 'B'])
+                self.frame.to_excel(path, 'test1', header=False)
+                self.frame.to_excel(path, 'test1', index=False)
+
+                # Test np.bool values read come back as float.
+                frame = (DataFrame([1, 0, True, False], dtype=np_type))
+                frame.to_excel(path, 'test1')
+                reader = ExcelFile(path)
+                recons = reader.parse('test1').astype(np_type)
+                tm.assert_frame_equal(frame, recons)
 
     def test_sheets(self):
         _skip_if_no_xlrd()