ENH: Added xlsxwriter as an ExcelWriter option.

ci/print_versions.py

@@ -92,6 +92,12 @@
 except:
     print("openpyxl: Not installed")
 
+try:
+    import xlsxwriter
+    print("xlsxwriter: %s" % xlsxwriter.__version__)
+except:
+    print("xlwt: Not installed")
+
 try:
     import xlrd
     print("xlrd: %s" % xlrd.__VERSION__)

ci/requirements-2.6.txt

@@ -4,3 +4,4 @@ python-dateutil==1.5
 pytz==2013b
 http://www.crummy.com/software/BeautifulSoup/bs4/download/4.2/beautifulsoup4-4.2.0.tar.gz
 html5lib==1.0b2
+xlsxwriter==0.4.2

ci/requirements-2.7.txt

@@ -8,6 +8,7 @@ numexpr==2.1
 tables==2.3.1
 matplotlib==1.1.1
 openpyxl==1.6.2
+xlsxwriter==0.4.2
 xlrd==0.9.2
 patsy==0.1.0
 html5lib==1.0b2

ci/requirements-2.7_LOCALE.txt

@@ -2,6 +2,7 @@ python-dateutil
 pytz==2013b
 xlwt==0.7.5
 openpyxl==1.6.2
+xlsxwriter==0.4.2
 xlrd==0.9.2
 numpy==1.6.1
 cython==0.19.1

ci/requirements-3.2.txt

@@ -1,6 +1,7 @@
 python-dateutil==2.1
 pytz==2013b
 openpyxl==1.6.2
+xlsxwriter==0.4.2
 xlrd==0.9.2
 numpy==1.6.2
 cython==0.19.1

ci/requirements-3.3.txt

@@ -1,6 +1,7 @@
 python-dateutil==2.1
 pytz==2013b
 openpyxl==1.6.2
+xlsxwriter==0.4.2
 xlrd==0.9.2
 html5lib==1.0b2
 numpy==1.7.1

doc/source/10min.rst

@@ -695,13 +695,13 @@ Writing to an excel file
 
 .. ipython:: python
 
-   df.to_excel('foo.xlsx', sheet_name='sheet1')
+   df.to_excel('foo.xlsx', sheet_name='Sheet1')
 
 Reading from an excel file
 
 .. ipython:: python
 
-   pd.read_excel('foo.xlsx', 'sheet1', index_col=None, na_values=['NA'])
+   pd.read_excel('foo.xlsx', 'Sheet1', index_col=None, na_values=['NA'])
 
 .. ipython:: python
    :suppress:

doc/source/io.rst

@@ -1667,7 +1667,7 @@ written.  For example:
 
 .. code-block:: python
 
-   df.to_excel('path_to_file.xlsx', sheet_name='sheet1')
+   df.to_excel('path_to_file.xlsx', sheet_name='Sheet1')
 
 Files with a ``.xls`` extension will be written using ``xlwt`` and those with
 a ``.xlsx`` extension will be written using ``openpyxl``.
@@ -1680,8 +1680,8 @@ one can use the ExcelWriter class, as in the following example:
 .. code-block:: python
 
    writer = ExcelWriter('path_to_file.xlsx')
-   df1.to_excel(writer, sheet_name='sheet1')
-   df2.to_excel(writer, sheet_name='sheet2')
+   df1.to_excel(writer, sheet_name='Sheet1')
+   df2.to_excel(writer, sheet_name='Sheet2')
    writer.save()
 
 .. _io.hdf5:

pandas/core/frame.py

@@ -1356,17 +1356,18 @@ def to_csv(self, path_or_buf, sep=",", na_rep='', float_format=None,
                                      tupleize_cols=tupleize_cols)
         formatter.save()
 
-    def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
+    def to_excel(self, excel_writer, sheet_name='Sheet1', na_rep='',
                  float_format=None, cols=None, header=True, index=True,
-                 index_label=None, startrow=0, startcol=0):
+                 index_label=None, startrow=0, startcol=0,
+                 engine=None):
         """
         Write DataFrame to a excel sheet
 
         Parameters
         ----------
         excel_writer : string or ExcelWriter object
             File path or existing ExcelWriter
-        sheet_name : string, default 'sheet1'
+        sheet_name : string, default 'Sheet1'
             Name of sheet which will contain DataFrame
         na_rep : string, default ''
             Missing data representation
@@ -1385,6 +1386,7 @@ def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
             sequence should be given if the DataFrame uses MultiIndex.
         startow : upper left cell row to dump data frame
         startcol : upper left cell column to dump data frame
+        engine : Excel writer class
 
 
         Notes
@@ -1393,14 +1395,14 @@ def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
         to the existing workbook.  This can be used to save different
         DataFrames to one workbook
         >>> writer = ExcelWriter('output.xlsx')
-        >>> df1.to_excel(writer,'sheet1')
-        >>> df2.to_excel(writer,'sheet2')
+        >>> df1.to_excel(writer,'Sheet1')
+        >>> df2.to_excel(writer,'Sheet2')
         >>> writer.save()
         """
         from pandas.io.excel import ExcelWriter
         need_save = False
         if isinstance(excel_writer, compat.string_types):
-            excel_writer = ExcelWriter(excel_writer)
+            excel_writer = ExcelWriter(excel_writer, engine)
             need_save = True
 
         formatter = fmt.ExcelFormatter(self,

pandas/core/panel.py

@@ -458,7 +458,7 @@ def to_sparse(self, fill_value=None, kind='block'):
                            default_kind=kind,
                            default_fill_value=fill_value)
 
-    def to_excel(self, path, na_rep=''):
+    def to_excel(self, path, na_rep='', engine=None):
         """
         Write each DataFrame in Panel to a separate excel sheet
 
@@ -468,9 +468,10 @@ def to_excel(self, path, na_rep=''):
             File path or existing ExcelWriter
         na_rep : string, default ''
             Missing data representation
+        engine : string, Excel writer class
         """
         from pandas.io.excel import ExcelWriter
-        writer = ExcelWriter(path)
+        writer = ExcelWriter(path, engine=engine)
         for item, df in compat.iteritems(self):
             name = str(item)
             df.to_excel(writer, name, na_rep=na_rep)

pandas/io/excel.py

@@ -12,9 +12,20 @@
 from pandas.tseries.period import Period
 from pandas import json
 from pandas.compat import map, zip, reduce, range, lrange
+from pandas.core import config
 import pandas.compat as compat
 from warnings import warn
 
+# Set up the io.excel specific configuration.
+writer_engine_doc = """
+: string
+    The default Excel engine. The options are 'openpyxl' (the default), 'xlwt'
+    and 'xlsxwriter'.
+"""
+with config.config_prefix('io.excel'):
+    config.register_option('writer_engine', None, writer_engine_doc,
+                            validator=str)
+
 
 def read_excel(path_or_buf, sheetname, **kwds):
     """Read an Excel table into a pandas DataFrame
@@ -256,7 +267,7 @@ def to_xls(style_dict, num_format_str=None):
         import xlwt
 
         def style_to_xlwt(item, firstlevel=True, field_sep=',', line_sep=';'):
-            """helper wich recursively generate an xlwt easy style string
+            """helper which recursively generate an xlwt easy style string
             for example:
 
               hstyle = {"font": {"bold": True},
@@ -318,6 +329,37 @@ def to_xlsx(style_dict):
 
         return xls_style
 
+    @staticmethod
+    def to_xlsxwriter(workbook, style_dict, num_format_str=None):
+        """
+        Converts a style_dict to an XlxsWriter format object.
+        Parameters
+        ----------
+        workbook:   Reference to the ExcelWriter XlxsWriter workbook.
+        style_dict: Style dictionary to convert.
+        num_format: Optional number format for the cell format.
+        """
+        if style_dict is None:
+            return None
+
+        # Create a XlsxWriter format object.
+        xl_format = workbook.add_format()
+
+        # Map the cell font to XlsxWriter font properties.
+        if style_dict.get('font'):
+            font = style_dict['font']
+            if font.get('bold'):
+                xl_format.set_bold()
+
+        # Map the cell borders to XlsxWriter border properties.
+        if style_dict.get('borders'):
+            xl_format.set_border()
+
+        if num_format_str is not None:
+            xl_format.set_num_format(num_format_str)
+
+        return xl_format
+
 
 def _conv_value(val):
     # convert value for excel dump
@@ -341,30 +383,35 @@ class ExcelWriter(object):
     path : string
         Path to xls file
     """
-    def __init__(self, path):
-        self.use_xlsx = True
-        if path.endswith('.xls'):
-            self.use_xlsx = False
-            import xlwt
-            self.book = xlwt.Workbook()
-            self.fm_datetime = xlwt.easyxf(
-                num_format_str='YYYY-MM-DD HH:MM:SS')
-            self.fm_date = xlwt.easyxf(num_format_str='YYYY-MM-DD')
-        else:
-            from openpyxl.workbook import Workbook
-            self.book = Workbook()  # optimized_write=True)
-            # open pyxl 1.6.1 adds a dummy sheet remove it
-            if self.book.worksheets:
-                self.book.remove_sheet(self.book.worksheets[0])
-        self.path = path
+    def __init__(self, path, engine=None, **engine_kwargs):
+
+        if engine is None:
+            default = config.get_option('io.excel.writer_engine')
+            if default is not None:
+                engine = default
+            elif path.endswith('.xls'):
+                engine = 'xlwt'
+            else:
+                engine = 'openpyxl'
+
+        try:
+            writer_init = getattr(self, "_init_%s" % engine)
+        except AttributeError:
+            raise ValueError("No engine: %s" % engine)
+
+        writer_init(path, **engine_kwargs)
+
         self.sheets = {}
         self.cur_sheet = None
 
     def save(self):
         """
         Save workbook to disk
         """
-        self.book.save(self.path)
+        if self.engine == 'xlsxwriter':
+            self.book.close()
+        else:
+            self.book.save(self.path)
 
     def write_cells(self, cells, sheet_name=None, startrow=0, startcol=0):
         """
@@ -381,16 +428,20 @@ def write_cells(self, cells, sheet_name=None, startrow=0, startcol=0):
         """
         if sheet_name is None:
             sheet_name = self.cur_sheet
+
         if sheet_name is None:  # pragma: no cover
             raise ValueError('Must pass explicit sheet_name or set '
-                             'cur_sheet property')
-        if self.use_xlsx:
-            self._writecells_xlsx(cells, sheet_name, startrow, startcol)
-        else:
-            self._writecells_xls(cells, sheet_name, startrow, startcol)
+                            'cur_sheet property')
 
-    def _writecells_xlsx(self, cells, sheet_name, startrow, startcol):
+        try:
+            _writecells = getattr(self, "_writecells_%s" % self.engine)
+        except AttributeError:
+            raise ValueError("No _writecells_%s() method" % self.engine)
 
+        _writecells(cells, sheet_name, startrow, startcol)
+
+    def _writecells_openpyxl(self, cells, sheet_name, startrow, startcol):
+        # Write the frame cells using openpyxl.
         from openpyxl.cell import get_column_letter
 
         if sheet_name in self.sheets:
@@ -426,7 +477,8 @@ def _writecells_xlsx(self, cells, sheet_name, startrow, startcol):
                                                cletterend,
                                                startrow + cell.mergestart + 1))
 
-    def _writecells_xls(self, cells, sheet_name, startrow, startcol):
+    def _writecells_xlwt(self, cells, sheet_name, startrow, startcol):
+        # Write the frame cells using xlwt.
         if sheet_name in self.sheets:
             wks = self.sheets[sheet_name]
         else:
@@ -464,3 +516,78 @@ def _writecells_xls(self, cells, sheet_name, startrow, startcol):
                 wks.write(startrow + cell.row,
                           startcol + cell.col,
                           val, style)
+
+    def _writecells_xlsxwriter(self, cells, sheet_name, startrow, startcol):
+        # Write the frame cells using xlsxwriter.
+        if sheet_name in self.sheets:
+            wks = self.sheets[sheet_name]
+        else:
+            wks = self.book.add_worksheet(sheet_name)
+            self.sheets[sheet_name] = wks
+
+        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"
+            if isinstance(cell.val, datetime.date):
+                num_format_str = "YYYY-MM-DD"
+
+            stylekey = json.dumps(cell.style)
+            if num_format_str:
+                stylekey += num_format_str
+
+            if stylekey in style_dict:
+                style = style_dict[stylekey]
+            else:
+                style = CellStyleConverter.to_xlsxwriter(self.book,
+                                                         cell.style,
+                                                         num_format_str)
+                style_dict[stylekey] = style
+
+            if cell.mergestart is not None and cell.mergeend is not None:
+                wks.merge_range(startrow + cell.row,
+                                startrow + cell.mergestart,
+                                startcol + cell.col,
+                                startcol + cell.mergeend,
+                                val, style)
+            else:
+                wks.write(startrow + cell.row,
+                          startcol + cell.col,
+                          val, style)
+
+    def _init_xlwt(self, filename, **engine_kwargs):
+        # Use the xlwt module as the Excel writer.
+        import xlwt
+
+        self.engine = 'xlwt'
+        self.path = filename
+        self.book = xlwt.Workbook()
+        self.fm_datetime = xlwt.easyxf(num_format_str='YYYY-MM-DD HH:MM:SS')
+        self.fm_date = xlwt.easyxf(num_format_str='YYYY-MM-DD')
+
+    def _init_openpyxl(self, filename, **engine_kwargs):
+        # Use the openpyxl module as the Excel writer.
+        from openpyxl.workbook import Workbook
+
+        self.engine = 'openpyxl'
+        self.path = filename
+        # Create workbook object with default optimized_write=True.
+        self.book = Workbook()
+        # Openpyxl 1.6.1 adds a dummy sheet. We remove it.
+        if self.book.worksheets:
+            self.book.remove_sheet(self.book.worksheets[0])
+
+    def _init_xlsxwriter(self, filename, **engine_kwargs):
+        # Use the xlsxwriter module as the Excel writer.
+        import xlsxwriter
+
+        options = dict(engine_kwargs)
+
+        options.setdefault('default_date_format', 'YYYY-MM-DD HH:MM:SS')
+
+        self.engine = 'xlsxwriter'
+        self.book = xlsxwriter.Workbook(filename, options)