ENH: Write excel comments, via styler.to_excel() tooltips (#58070)

Co-authored-by: diogomsmiranda <[email protected]>

Author: Dacops

doc/source/whatsnew/v3.0.0.rst

@@ -39,6 +39,7 @@ Other enhancements
 - Users can globally disable any ``PerformanceWarning`` by setting the option ``mode.performance_warnings`` to ``False`` (:issue:`56920`)
 - :meth:`Styler.format_index_names` can now be used to format the index and column names (:issue:`48936` and :issue:`47489`)
 - :class:`.errors.DtypeWarning` improved to include column names when mixed data types are detected (:issue:`58174`)
+- :func:`DataFrame.to_excel` now supports writing notes to an excel files via :meth:`Styler.set_tooltips` (:issue:`58070`)
 - :meth:`DataFrame.corrwith` now accepts ``min_periods`` as optional arguments, as in :meth:`DataFrame.corr` and :meth:`Series.corr` (:issue:`9490`)
 - :meth:`DataFrame.cummin`, :meth:`DataFrame.cummax`, :meth:`DataFrame.cumprod` and :meth:`DataFrame.cumsum` methods now have a ``numeric_only`` parameter (:issue:`53072`)
 - :meth:`DataFrame.fillna` and :meth:`Series.fillna` can now accept ``value=None``; for non-object dtype the corresponding NA value will be used (:issue:`57723`)

pandas/core/generic.py

@@ -2127,10 +2127,11 @@ def _repr_data_resource_(self):
         storage_options_versionadded="1.2.0",
         extra_parameters=textwrap.dedent(
             """\
-        engine_kwargs : dict, optional
-            Arbitrary keyword arguments passed to excel engine.
-    """
+            engine_kwargs : dict, optional
+                Arbitrary keyword arguments passed to excel engine.
+            """
         ),
+        extra_examples="",
     )
     def to_excel(
         self,
@@ -2262,6 +2263,7 @@ def to_excel(
         automatically chosen depending on the file extension):
 
         >>> df1.to_excel("output1.xlsx", engine="xlsxwriter")  # doctest: +SKIP
+        {extra_examples}
         """
         if engine_kwargs is None:
             engine_kwargs = {}

pandas/io/excel/_base.py

@@ -1212,6 +1212,7 @@ def _write_cells(
         startrow: int = 0,
         startcol: int = 0,
         freeze_panes: tuple[int, int] | None = None,
+        notes: DataFrame | None = None,
     ) -> None:
         """
         Write given formatted cells into Excel an excel sheet
@@ -1220,6 +1221,8 @@ def _write_cells(
         ----------
         cells : generator
             cell of formatted data to save to Excel sheet
+        notes: DataFrame
+            DataFrame containing notes to be written to the Excel sheet
         sheet_name : str, default None
             Name of Excel sheet, if None, then use self.cur_sheet
         startrow : upper left cell row to dump data frame

pandas/io/excel/_odswriter.py

@@ -27,6 +27,8 @@
         WriteExcelBuffer,
     )
 
+    from pandas.core.frame import DataFrame
+
     from pandas.io.formats.excel import ExcelCell
 
 
@@ -99,6 +101,7 @@ def _write_cells(
         startrow: int = 0,
         startcol: int = 0,
         freeze_panes: tuple[int, int] | None = None,
+        notes: DataFrame | None = None,
     ) -> None:
         """
         Write the frame cells using odf
@@ -110,6 +113,14 @@ def _write_cells(
         )
         from odf.text import P
 
+        if notes is not None:
+            raise NotImplementedError(
+                """
+                Notes are not supported by the odswriter engine,
+                see https://github.com/mmulqueen/odswriter
+                """
+            )
+
         sheet_name = self._get_sheet_name(sheet_name)
         assert sheet_name is not None
 

pandas/io/excel/_openpyxl.py

@@ -37,6 +37,8 @@
         WriteExcelBuffer,
     )
 
+    from pandas.core.frame import DataFrame
+
 
 class OpenpyxlWriter(ExcelWriter):
     _engine = "openpyxl"
@@ -447,7 +449,10 @@ def _write_cells(
         startrow: int = 0,
         startcol: int = 0,
         freeze_panes: tuple[int, int] | None = None,
+        notes: DataFrame | None = None,
     ) -> None:
+        from openpyxl.comments import Comment
+
         # Write the frame cells using openpyxl.
         sheet_name = self._get_sheet_name(sheet_name)
 
@@ -484,7 +489,11 @@ def _write_cells(
                 row=freeze_panes[0] + 1, column=freeze_panes[1] + 1
             )
 
+        notes_col = None
+
         for cell in cells:
+            if notes_col is None:
+                notes_col = startcol + cell.col + 1
             xcell = wks.cell(
                 row=startrow + cell.row + 1, column=startcol + cell.col + 1
             )
@@ -530,6 +539,20 @@ def _write_cells(
                             for k, v in style_kwargs.items():
                                 setattr(xcell, k, v)
 
+        if notes is None or notes_col is None:
+            return
+
+        for row_idx, (_, val) in enumerate(notes.iterrows()):
+            for col_idx, note in enumerate(val):
+                xcell = wks.cell(
+                    # first row has columns and openpyxl starts counting at 1, not 0
+                    row=row_idx + 2,
+                    column=col_idx + notes_col,  # n columns with indexes
+                )
+                if note:
+                    comment = Comment(note, "")
+                    xcell.comment = comment
+
 
 class OpenpyxlReader(BaseExcelReader["Workbook"]):
     @doc(storage_options=_shared_docs["storage_options"])

pandas/io/excel/_xlsxwriter.py

@@ -20,6 +20,8 @@
         WriteExcelBuffer,
     )
 
+    from pandas.core.frame import DataFrame
+
 
 class _XlsxStyler:
     # Map from openpyxl-oriented styles to flatter xlsxwriter representation
@@ -245,6 +247,7 @@ def _write_cells(
         startrow: int = 0,
         startcol: int = 0,
         freeze_panes: tuple[int, int] | None = None,
+        notes: DataFrame | None = None,
     ) -> None:
         # Write the frame cells using xlsxwriter.
         sheet_name = self._get_sheet_name(sheet_name)
@@ -258,6 +261,8 @@ def _write_cells(
         if validate_freeze_panes(freeze_panes):
             wks.freeze_panes(*(freeze_panes))
 
+        notes_col = None
+
         for cell in cells:
             val, fmt = self._value_with_fmt(cell.val)
 
@@ -281,4 +286,19 @@ def _write_cells(
                     style,
                 )
             else:
+                if notes_col is None:
+                    notes_col = startcol + cell.col
                 wks.write(startrow + cell.row, startcol + cell.col, val, style)
+
+        if notes is None or notes_col is None:
+            return
+
+        for row_idx, (_, row) in enumerate(notes.iterrows()):
+            for col_idx, note in enumerate(row):
+                if note == "":
+                    continue
+                wks.write_comment(
+                    row_idx + 1,  # first row has columns
+                    col_idx + notes_col,  # n columns with indexes
+                    str(note),
+                )

pandas/io/formats/excel.py

@@ -553,7 +553,10 @@ def __init__(
     ) -> None:
         self.rowcounter = 0
         self.na_rep = na_rep
+        self.notes = None
         if not isinstance(df, DataFrame):
+            if df.tooltips is not None:
+                self.notes = df.tooltips.tt_data
             self.styler = df
             self.styler._compute()  # calculate applied styles
             df = df.data
@@ -945,6 +948,7 @@ def write(
                 startrow=startrow,
                 startcol=startcol,
                 freeze_panes=freeze_panes,
+                notes=self.notes,
             )
         finally:
             # make sure to close opened file handles

pandas/io/formats/style.py

@@ -82,6 +82,7 @@
 
     from pandas import ExcelWriter
 
+import textwrap
 
 ####
 # Shared Doc Strings
@@ -537,6 +538,19 @@ def set_tooltips(
         storage_options=_shared_docs["storage_options"],
         storage_options_versionadded="1.5.0",
         extra_parameters="",
+        extra_examples=textwrap.dedent(
+            """\
+            If you wish to write excel notes to the workbook, you can do so by
+            passing a DataFrame to ``set_tooltips``. This process is independent
+            from writing data to the workbook, therefore both DataFrames can have
+            different dimensions.
+
+            >>> notes = pd.DataFrame(
+            ...     [["cell 1", "cell 2"], ["cell 3", "cell 4"]],
+            ... )  # doctest: +SKIP
+            >>> df1.style.set_tooltips(notes).to_excel("output.xlsx")  # doctest: +SKIP
+            """
+        ),
     )
     def to_excel(
         self,