Deprecate using xlrd engine and change default engine to read excel…

… files to openpyxl

doc/source/whatsnew/v1.1.0.rst

@@ -820,6 +820,7 @@ Deprecations
   precision through the ``rtol``, and ``atol`` parameters, thus deprecating the
   ``check_less_precise`` parameter. (:issue:`13357`).
 - :func:`DataFrame.melt` accepting a value_name that already exists is deprecated, and will be removed in a future version (:issue:`34731`)
+- :func:`read_excel` default engine "xlrd" is replaced by "openpyxl" because "xlrd" is deprecated (:issue:`28547`).
 
 .. ---------------------------------------------------------------------------
 

pandas/io/excel/_base.py

@@ -4,6 +4,7 @@
 import os
 from textwrap import fill
 from typing import Union
+import warnings
 
 from pandas._config import config
 
@@ -810,8 +811,7 @@ def _is_ods_stream(stream: Union[BufferedIOBase, RawIOBase]) -> bool:
 class ExcelFile:
     """
     Class for parsing tabular excel sheets into DataFrame objects.
-
-    Uses xlrd engine by default. See read_excel for more documentation
+    Uses xlrd, openpyxl or odf. See read_excel for more documentation
 
     Parameters
     ----------
@@ -822,7 +822,7 @@ class ExcelFile:
     engine : str, default None
         If io is not a buffer or path, this must be set to identify io.
         Supported engines: ``xlrd``, ``openpyxl``, ``odf``, ``pyxlsb``,
-        default ``xlrd``.
+        default ``openpyxl``, ``xlrd`` for .xls files, ``odf`` for .ods files.
         Engine compatibility :
         - ``xlrd`` supports most old/new Excel file formats.
         - ``openpyxl`` supports newer Excel file formats.
@@ -844,14 +844,24 @@ class ExcelFile:
 
     def __init__(self, path_or_buffer, engine=None):
         if engine is None:
-            engine = "xlrd"
+            engine = "openpyxl"
             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"
+                elif ext == ".xls":
+                    engine = "xlrd"
+
+        elif engine == "xlrd":
+            warnings.warn(
+                'The Excel reader engine "xlrd" is deprecated, use "openpyxl" instead. '
+                'Specify engine="openpyxl" to suppress this warning.',
+                FutureWarning,
+                stacklevel=2,
+            )
         if engine not in self._engines:
             raise ValueError(f"Unknown engine: {engine}")
 

pandas/io/excel/_openpyxl.py

@@ -1,3 +1,4 @@
+from datetime import datetime
 from typing import List
 
 import numpy as np
@@ -511,7 +512,11 @@ def _convert_cell(self, cell, convert_float: bool) -> Scalar:
 
         # TODO: replace with openpyxl constants
         if cell.is_date:
-            return cell.value
+            try:
+                # workaround for inaccurate timestamp notation in excel
+                return datetime.fromtimestamp(round(cell.value.timestamp()))
+            except (AttributeError, OSError):
+                return cell.value
         elif cell.data_type == "e":
             return np.nan
         elif cell.data_type == "b":

pandas/tests/io/excel/test_readers.py

@@ -40,6 +40,9 @@ def ignore_xlrd_time_clock_warning():
         marks=[
             td.skip_if_no("xlrd"),
             pytest.mark.filterwarnings("ignore:.*(tree\\.iter|html argument)"),
+            pytest.mark.filterwarnings(
+                'ignore:The Excel reader engine "xlrd" is deprecated,'
+            ),
         ],
     ),
     pytest.param(
@@ -52,8 +55,8 @@ def ignore_xlrd_time_clock_warning():
     pytest.param(
         None,
         marks=[
-            td.skip_if_no("xlrd"),
-            pytest.mark.filterwarnings("ignore:.*(tree\\.iter|html argument)"),
+            td.skip_if_no("openpyxl"),
+            pytest.mark.filterwarnings("ignore:.*html argument"),
         ],
     ),
     pytest.param("pyxlsb", marks=td.skip_if_no("pyxlsb")),
@@ -69,6 +72,8 @@ def _is_valid_engine_ext_pair(engine, read_ext: str) -> bool:
     engine = engine.values[0]
     if engine == "openpyxl" and read_ext == ".xls":
         return False
+    if engine is None and read_ext == ".xls":
+        return False
     if engine == "odf" and read_ext != ".ods":
         return False
     if read_ext == ".ods" and engine != "odf":
@@ -579,7 +584,7 @@ def test_date_conversion_overflow(self, read_ext):
             columns=["DateColWithBigInt", "StringCol"],
         )
 
-        if pd.read_excel.keywords["engine"] == "openpyxl":
+        if pd.read_excel.keywords["engine"] in ["openpyxl", None]:
             pytest.xfail("Maybe not supported by openpyxl")
 
         result = pd.read_excel("testdateoverflow" + read_ext)
@@ -962,12 +967,28 @@ def test_read_excel_squeeze(self, read_ext):
         expected = pd.Series([1, 2, 3], name="a")
         tm.assert_series_equal(actual, expected)
 
-    def test_deprecated_kwargs(self, read_ext):
+    def test_deprecated_kwargs(self, engine, read_ext):
+        if engine == "xlrd":
+            pytest.skip("Use of xlrd engine produces a FutureWarning as well")
+
         with tm.assert_produces_warning(FutureWarning, raise_on_extra_warnings=False):
             pd.read_excel("test1" + read_ext, "Sheet1", 0)
 
         pd.read_excel("test1" + read_ext)
 
+    def test_excel_high_surrogate(self, engine, read_ext):
+        # GH 23809
+        if read_ext != ".xlsx":
+            pytest.skip("Test is only applicable to .xlsx file")
+        if engine in ["openpyxl", None]:
+            pytest.skip("Test does not work for openpyxl")
+
+        expected = pd.DataFrame(["\udc88"], columns=["Column1"])
+
+        # should not produce a segmentation violation
+        actual = pd.read_excel("high_surrogate.xlsx")
+        tm.assert_frame_equal(expected, actual)
+
 
 class TestExcelFileRead:
     @pytest.fixture(autouse=True)
@@ -1123,14 +1144,6 @@ def test_excel_read_binary(self, engine, read_ext):
         actual = pd.read_excel(data, engine=engine)
         tm.assert_frame_equal(expected, actual)
 
-    def test_excel_high_surrogate(self, engine):
-        # GH 23809
-        expected = pd.DataFrame(["\udc88"], columns=["Column1"])
-
-        # should not produce a segmentation violation
-        actual = pd.read_excel("high_surrogate.xlsx")
-        tm.assert_frame_equal(expected, actual)
-
     @pytest.mark.parametrize("filename", ["df_empty.xlsx", "df_equals.xlsx"])
     def test_header_with_index_col(self, engine, filename):
         # GH 33476

pandas/tests/io/excel/test_writers.py

@@ -351,12 +351,16 @@ def test_excel_sheet_by_name_raise(self, path, engine):
             msg = "sheet 0 not found"
             with pytest.raises(ValueError, match=msg):
                 pd.read_excel(xl, "0")
-        else:
+        elif engine == "xlwt":
             import xlrd
 
             msg = "No sheet named <'0'>"
             with pytest.raises(xlrd.XLRDError, match=msg):
                 pd.read_excel(xl, sheet_name="0")
+        else:  # openpyxl
+            msg = "Worksheet 0 does not exist."
+            with pytest.raises(KeyError, match=msg):
+                pd.read_excel(xl, sheet_name="0")
 
     def test_excel_writer_context_manager(self, frame, path):
         with ExcelWriter(path) as writer:
@@ -1199,6 +1203,9 @@ def test_datetimes(self, path):
 
         tm.assert_series_equal(write_frame["A"], read_frame["A"])
 
+    @pytest.mark.filterwarnings(
+        'ignore:The Excel reader engine "xlrd" is deprecated:FutureWarning'
+    )
     def test_bytes_io(self, engine):
         # see gh-7074
         bio = BytesIO()
@@ -1209,8 +1216,15 @@ def test_bytes_io(self, engine):
         df.to_excel(writer)
         writer.save()
 
+        if engine == "xlwt":
+            read_engine = "xlrd"
+        elif engine == "xlsxwriter":
+            read_engine = "openpyxl"
+        else:
+            read_engine = engine
+
         bio.seek(0)
-        reread_df = pd.read_excel(bio, index_col=0)
+        reread_df = pd.read_excel(bio, index_col=0, engine=read_engine)
         tm.assert_frame_equal(df, reread_df)
 
     def test_write_lists_dict(self, path):

pandas/tests/io/excel/test_xlrd.py

@@ -17,6 +17,9 @@ def skip_ods_and_xlsb_files(read_ext):
         pytest.skip("Not valid for xlrd")
 
 
+@pytest.mark.filterwarnings(
+    'ignore:The Excel reader engine "xlrd" is deprecated:FutureWarning'
+)
 def test_read_xlrd_book(read_ext, frame):
     df = frame
 
@@ -36,8 +39,31 @@ def test_read_xlrd_book(read_ext, frame):
 
 
 # TODO: test for openpyxl as well
+@pytest.mark.filterwarnings(
+    'ignore:The Excel reader engine "xlrd" is deprecated:FutureWarning'
+)
 def test_excel_table_sheet_by_index(datapath, read_ext):
     path = datapath("io", "data", "excel", f"test1{read_ext}")
-    with pd.ExcelFile(path) as excel:
+    with pd.ExcelFile(path, engine="xlrd") as excel:
         with pytest.raises(xlrd.XLRDError):
             pd.read_excel(excel, sheet_name="asdf")
+
+
+def test_excel_file_warning_with_xlsx_file(datapath):
+    # GH 29375
+    path = datapath("io", "data", "excel", "test1.xlsx")
+    with tm.assert_produces_warning(
+        FutureWarning, check_stacklevel=True, raise_on_extra_warnings=False
+    ) as w:
+        pd.ExcelFile(path, engine="xlrd")
+        assert '"xlrd" is deprecated, use "openpyxl" instead.' in str(w[0].message)
+
+
+def test_read_excel_warning_with_xlsx_file(tmpdir, datapath):
+    # GH 29375
+    path = datapath("io", "data", "excel", "test1.xlsx")
+    with tm.assert_produces_warning(
+        FutureWarning, check_stacklevel=False, raise_on_extra_warnings=False
+    ) as w:
+        pd.read_excel(path, "Sheet1", engine="xlrd")
+        assert '"xlrd" is deprecated, use "openpyxl" instead.' in str(w[0].message)