ENH: Add ods writer

doc/source/whatsnew/v1.1.0.rst

@@ -316,6 +316,7 @@ Other enhancements
 - :meth:`~pandas.io.gbq.read_gbq` now supports the ``max_results`` kwarg from ``pandas-gbq`` (:issue:`34639`).
 - :meth:`Dataframe.cov` and :meth:`Series.cov` now support a new parameter ddof to support delta degrees of freedom as in the corresponding numpy methods (:issue:`34611`).
 - :meth:`DataFrame.to_html` and :meth:`DataFrame.to_string`'s ``col_space`` parameter now accepts a list of dict to change only some specific columns' width (:issue:`28917`).
+- :meth:`DataFrame.to_excel` can now also write OpenOffice spreadsheet (.ods) files (:issue:`27222`)
 
 .. ---------------------------------------------------------------------------
 
@@ -1018,6 +1019,7 @@ I/O
 - Bug in :meth:`~SQLDatabase.execute` was raising a ``ProgrammingError`` for some DB-API drivers when the SQL statement contained the `%` character and no parameters were present (:issue:`34211`)
 - Bug in :meth:`~pandas.io.stata.StataReader` which resulted in categorical variables with difference dtypes when reading data using an iterator. (:issue:`31544`)
 - :meth:`HDFStore.keys` has now an optional `include` parameter that allows the retrieval of all native HDF5 table names (:issue:`29916`)
+- Bug in :meth:`read_excel` for ODS files removes 0.0 values (:issue:`27222`)
 
 Plotting
 ^^^^^^^^

pandas/core/config_init.py

@@ -553,6 +553,7 @@ def use_inf_as_na_cb(key):
 _xls_options = ["xlwt"]
 _xlsm_options = ["openpyxl"]
 _xlsx_options = ["openpyxl", "xlsxwriter"]
+_ods_options = ["odf"]
 
 
 with cf.config_prefix("io.excel.xls"):
@@ -581,6 +582,15 @@ def use_inf_as_na_cb(key):
     )
 
 
+with cf.config_prefix("io.excel.ods"):
+    cf.register_option(
+        "writer",
+        "auto",
+        writer_engine_doc.format(ext="ods", others=", ".join(_ods_options)),
+        validator=str,
+    )
+
+
 # Set up the io.parquet specific configuration.
 parquet_engine_doc = """
 : string

pandas/io/excel/__init__.py

@@ -1,4 +1,5 @@
 from pandas.io.excel._base import ExcelFile, ExcelWriter, read_excel
+from pandas.io.excel._odswriter import _ODSWriter
 from pandas.io.excel._openpyxl import _OpenpyxlWriter
 from pandas.io.excel._util import register_writer
 from pandas.io.excel._xlsxwriter import _XlsxWriter
@@ -14,3 +15,6 @@
 
 
 register_writer(_XlsxWriter)
+
+
+register_writer(_ODSWriter)

pandas/io/excel/_base.py

@@ -1,8 +1,9 @@
 import abc
 import datetime
-from io import BytesIO
+from io import BufferedIOBase, BytesIO, RawIOBase
 import os
 from textwrap import fill
+from typing import Union
 
 from pandas._config import config
 
@@ -533,13 +534,13 @@ class ExcelWriter(metaclass=abc.ABCMeta):
     """
     Class for writing DataFrame objects into excel sheets.
 
-    Default is to use xlwt for xls, openpyxl for xlsx.
+    Default is to use xlwt for xls, openpyxl for xlsx, odf for ods.
     See DataFrame.to_excel for typical usage.
 
     Parameters
     ----------
     path : str
-        Path to xls or xlsx file.
+        Path to xls or xlsx or ods file.
     engine : str (optional)
         Engine to use for writing. If None, defaults to
         ``io.excel.<extension>.writer``.  NOTE: can only be passed as a keyword
@@ -692,10 +693,7 @@ def __init__(
         # validate that this engine can handle the extension
         if isinstance(path, str):
             ext = os.path.splitext(path)[-1]
-        else:
-            ext = "xls" if engine == "xlwt" else "xlsx"
-
-        self.check_extension(ext)
+            self.check_extension(ext)
 
         self.path = path
         self.sheets = {}
@@ -781,6 +779,34 @@ def close(self):
         return self.save()
 
 
+def _is_ods_stream(stream: Union[BufferedIOBase, RawIOBase]) -> bool:
+    """
+    Check if the stream is an OpenDocument Spreadsheet (.ods) file
+
+    It uses magic values inside the stream
+
+    Parameters
+    ----------
+    stream : Union[BufferedIOBase, RawIOBase]
+        IO stream with data which might be an ODS file
+
+    Returns
+    -------
+    is_ods : bool
+        Boolean indication that this is indeed an ODS file or not
+    """
+    stream.seek(0)
+    is_ods = False
+    if stream.read(4) == b"PK\003\004":
+        stream.seek(30)
+        is_ods = (
+            stream.read(54) == b"mimetype"
+            b"application/vnd.oasis.opendocument.spreadsheet"
+        )
+    stream.seek(0)
+    return is_ods
+
+
 class ExcelFile:
     """
     Class for parsing tabular excel sheets into DataFrame objects.
@@ -789,8 +815,8 @@ class ExcelFile:
 
     Parameters
     ----------
-    io : str, path object (pathlib.Path or py._path.local.LocalPath),
-         a file-like object, xlrd workbook or openpypl workbook.
+    path_or_buffer : str, path object (pathlib.Path or py._path.local.LocalPath),
+        a file-like object, xlrd workbook or openpypl workbook.
         If a string or path object, expected to be a path to a
         .xls, .xlsx, .xlsb, .xlsm, .odf, .ods, or .odt file.
     engine : str, default None
@@ -816,18 +842,25 @@ class ExcelFile:
         "pyxlsb": _PyxlsbReader,
     }
 
-    def __init__(self, io, engine=None):
+    def __init__(self, path_or_buffer, engine=None):
         if engine is None:
             engine = "xlrd"
+            if isinstance(path_or_buffer, (BufferedIOBase, RawIOBase)):
+                if _is_ods_stream(path_or_buffer):
+                    engine = "odf"
+            else:
+                ext = os.path.splitext(str(path_or_buffer))[-1]
+                if ext == ".ods":
+                    engine = "odf"
         if engine not in self._engines:
             raise ValueError(f"Unknown engine: {engine}")
 
         self.engine = engine
 
         # Could be a str, ExcelFile, Book, etc.
-        self.io = io
+        self.io = path_or_buffer
         # Always a string
-        self._io = stringify_path(io)
+        self._io = stringify_path(path_or_buffer)
 
         self._reader = self._engines[engine](self._io)
 

pandas/io/excel/_odfreader.py

@@ -1,5 +1,7 @@
 from typing import List, cast
 
+import numpy as np
+
 from pandas._typing import FilePathOrBuffer, Scalar
 from pandas.compat._optional import import_optional_dependency
 
@@ -148,6 +150,9 @@ def _is_empty_row(self, row) -> bool:
     def _get_cell_value(self, cell, convert_float: bool) -> Scalar:
         from odf.namespaces import OFFICENS
 
+        if str(cell) == "#N/A":
+            return np.nan
+
         cell_type = cell.attributes.get((OFFICENS, "value-type"))
         if cell_type == "boolean":
             if str(cell) == "TRUE":
@@ -158,10 +163,6 @@ def _get_cell_value(self, cell, convert_float: bool) -> Scalar:
         elif cell_type == "float":
             # GH5394
             cell_value = float(cell.attributes.get((OFFICENS, "value")))
-
-            if cell_value == 0.0:  # NA handling
-                return str(cell)
-
             if convert_float:
                 val = int(cell_value)
                 if val == cell_value: