From 54cd6d7406ee9431d45e8bcd1b0ae53389cfb56c Mon Sep 17 00:00:00 2001
From: glypt <63284048+glypt@users.noreply.github.com>
Date: Thu, 27 Nov 2025 16:25:32 +0100
Subject: [PATCH] fix: do not consider singleton cells in xlsx as TableItems
 but rather TextItems (#2589)

fix: do not handle 1x1 cell as a tableitem but as a textitem

Signed-off-by: glypt <8trash-can8@protonmail.ch>
---
 docling/backend/msexcel_backend.py            | 116 ++--
 docling/datamodel/backend_options.py          |  14 +
 .../xlsx_05_table_with_title.xlsx.itxt        |   4 +
 .../xlsx_05_table_with_title.xlsx.json        | 578 ++++++++++++++++++
 .../xlsx_05_table_with_title.xlsx.md          |  11 +
 tests/data/xlsx/xlsx_05_table_with_title.xlsx | Bin 0 -> 6335 bytes
 tests/test_backend_msexcel.py                 |  46 +-
 7 files changed, 729 insertions(+), 40 deletions(-)
 create mode 100644 tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.itxt
 create mode 100644 tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.json
 create mode 100644 tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.md
 create mode 100644 tests/data/xlsx/xlsx_05_table_with_title.xlsx

diff --git a/docling/backend/msexcel_backend.py b/docling/backend/msexcel_backend.py
index aeaed4f1..2bcc34d7 100644
--- a/docling/backend/msexcel_backend.py
+++ b/docling/backend/msexcel_backend.py
@@ -8,6 +8,7 @@ from docling_core.types.doc import (
     ContentLayer,
     CoordOrigin,
     DocItem,
+    DocItemLabel,
     DoclingDocument,
     DocumentOrigin,
     GroupLabel,
@@ -31,6 +32,7 @@ from docling.backend.abstract_backend import (
     DeclarativeDocumentBackend,
     PaginatedDocumentBackend,
 )
+from docling.datamodel.backend_options import MsExcelBackendOptions
 from docling.datamodel.base_models import InputFormat
 from docling.datamodel.document import InputDocument
 
@@ -116,18 +118,22 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
 
     @override
     def __init__(
-        self, in_doc: "InputDocument", path_or_stream: Union[BytesIO, Path]
+        self,
+        in_doc: "InputDocument",
+        path_or_stream: Union[BytesIO, Path],
+        options: MsExcelBackendOptions = MsExcelBackendOptions(),
     ) -> None:
         """Initialize the MsExcelDocumentBackend object.
 
         Parameters:
             in_doc: The input document object.
             path_or_stream: The path or stream to the Excel file.
+            options: Backend options for Excel parsing.
 
         Raises:
             RuntimeError: An error occurred parsing the file.
         """
-        super().__init__(in_doc, path_or_stream)
+        super().__init__(in_doc, path_or_stream, options)
 
         # Initialise the parents for the hierarchy
         self.max_levels = 10
@@ -277,51 +283,83 @@ class MsExcelDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentBacken
             content_layer = self._get_sheet_content_layer(sheet)
             tables = self._find_data_tables(sheet)
 
+            treat_singleton_as_text = (
+                isinstance(self.options, MsExcelBackendOptions)
+                and self.options.treat_singleton_as_text
+            )
+
             for excel_table in tables:
                 origin_col = excel_table.anchor[0]
                 origin_row = excel_table.anchor[1]
                 num_rows = excel_table.num_rows
                 num_cols = excel_table.num_cols
 
-                table_data = TableData(
-                    num_rows=num_rows,
-                    num_cols=num_cols,
-                    table_cells=[],
-                )
-
-                for excel_cell in excel_table.data:
-                    cell = TableCell(
-                        text=excel_cell.text,
-                        row_span=excel_cell.row_span,
-                        col_span=excel_cell.col_span,
-                        start_row_offset_idx=excel_cell.row,
-                        end_row_offset_idx=excel_cell.row + excel_cell.row_span,
-                        start_col_offset_idx=excel_cell.col,
-                        end_col_offset_idx=excel_cell.col + excel_cell.col_span,
-                        column_header=excel_cell.row == 0,
-                        row_header=False,
-                    )
-                    table_data.table_cells.append(cell)
-
-                page_no = self.workbook.index(sheet) + 1
-                doc.add_table(
-                    data=table_data,
-                    parent=self.parents[0],
-                    prov=ProvenanceItem(
-                        page_no=page_no,
-                        charspan=(0, 0),
-                        bbox=BoundingBox.from_tuple(
-                            (
-                                origin_col,
-                                origin_row,
-                                origin_col + num_cols,
-                                origin_row + num_rows,
+                if (
+                    treat_singleton_as_text
+                    and num_rows == 1
+                    and num_cols == 1
+                    and excel_table.data
+                ):
+                    page_no = self.workbook.index(sheet) + 1
+                    doc.add_text(
+                        text=excel_table.data[0].text,
+                        label=DocItemLabel.TEXT,
+                        parent=self.parents[0],
+                        prov=ProvenanceItem(
+                            page_no=page_no,
+                            charspan=(0, 0),
+                            bbox=BoundingBox.from_tuple(
+                                (
+                                    origin_col,
+                                    origin_row,
+                                    origin_col + num_cols,
+                                    origin_row + num_rows,
+                                ),
+                                origin=CoordOrigin.TOPLEFT,
                             ),
-                            origin=CoordOrigin.TOPLEFT,
                         ),
-                    ),
-                    content_layer=content_layer,
-                )
+                        content_layer=content_layer,
+                    )
+                else:
+                    table_data = TableData(
+                        num_rows=num_rows,
+                        num_cols=num_cols,
+                        table_cells=[],
+                    )
+
+                    for excel_cell in excel_table.data:
+                        cell = TableCell(
+                            text=excel_cell.text,
+                            row_span=excel_cell.row_span,
+                            col_span=excel_cell.col_span,
+                            start_row_offset_idx=excel_cell.row,
+                            end_row_offset_idx=excel_cell.row + excel_cell.row_span,
+                            start_col_offset_idx=excel_cell.col,
+                            end_col_offset_idx=excel_cell.col + excel_cell.col_span,
+                            column_header=excel_cell.row == 0,
+                            row_header=False,
+                        )
+                        table_data.table_cells.append(cell)
+
+                    page_no = self.workbook.index(sheet) + 1
+                    doc.add_table(
+                        data=table_data,
+                        parent=self.parents[0],
+                        prov=ProvenanceItem(
+                            page_no=page_no,
+                            charspan=(0, 0),
+                            bbox=BoundingBox.from_tuple(
+                                (
+                                    origin_col,
+                                    origin_row,
+                                    origin_col + num_cols,
+                                    origin_row + num_rows,
+                                ),
+                                origin=CoordOrigin.TOPLEFT,
+                            ),
+                        ),
+                        content_layer=content_layer,
+                    )
 
         return doc
 
diff --git a/docling/datamodel/backend_options.py b/docling/datamodel/backend_options.py
index 594be689..ea5cf70c 100644
--- a/docling/datamodel/backend_options.py
+++ b/docling/datamodel/backend_options.py
@@ -71,12 +71,26 @@ class PdfBackendOptions(BaseBackendOptions):
     password: Optional[SecretStr] = None
 
 
+class MsExcelBackendOptions(BaseBackendOptions):
+    """Options specific to the MS Excel backend."""
+
+    kind: Literal["xlsx"] = Field("xlsx", exclude=True, repr=False)
+    treat_singleton_as_text: bool = Field(
+        False,
+        description=(
+            "Whether to treat singleton cells (1x1 tables with empty neighboring "
+            "cells) as TextItem instead of TableItem."
+        ),
+    )
+
+
 BackendOptions = Annotated[
     Union[
         DeclarativeBackendOptions,
         HTMLBackendOptions,
         MarkdownBackendOptions,
         PdfBackendOptions,
+        MsExcelBackendOptions,
     ],
     Field(discriminator="kind"),
 ]
diff --git a/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.itxt b/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.itxt
new file mode 100644
index 00000000..30e0db58
--- /dev/null
+++ b/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.itxt
@@ -0,0 +1,4 @@
+item-0 at level 0: unspecified: group _root_
+  item-1 at level 1: section: group sheet: Duck Observations
+    item-2 at level 2: table with [1x1]
+    item-3 at level 2: table with [7x2]
\ No newline at end of file
diff --git a/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.json b/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.json
new file mode 100644
index 00000000..04441a9f
--- /dev/null
+++ b/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.json
@@ -0,0 +1,578 @@
+{
+  "schema_name": "DoclingDocument",
+  "version": "1.7.0",
+  "name": "xlsx_05_table_with_title",
+  "origin": {
+    "mimetype": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+    "binary_hash": 18126553641942797758,
+    "filename": "xlsx_05_table_with_title.xlsx",
+    "uri": null
+  },
+  "furniture": {
+    "self_ref": "#/furniture",
+    "parent": null,
+    "children": [],
+    "content_layer": "furniture",
+    "name": "_root_",
+    "label": "unspecified"
+  },
+  "body": {
+    "self_ref": "#/body",
+    "parent": null,
+    "children": [
+      {
+        "cref": "#/groups/0"
+      }
+    ],
+    "content_layer": "body",
+    "name": "_root_",
+    "label": "unspecified"
+  },
+  "groups": [
+    {
+      "self_ref": "#/groups/0",
+      "parent": {
+        "cref": "#/body"
+      },
+      "children": [
+        {
+          "cref": "#/tables/0"
+        },
+        {
+          "cref": "#/tables/1"
+        }
+      ],
+      "content_layer": "body",
+      "name": "sheet: Duck Observations",
+      "label": "section"
+    }
+  ],
+  "texts": [],
+  "pictures": [],
+  "tables": [
+    {
+      "self_ref": "#/tables/0",
+      "parent": {
+        "cref": "#/groups/0"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "table",
+      "prov": [
+        {
+          "page_no": 1,
+          "bbox": {
+            "l": 1.0,
+            "t": 1.0,
+            "r": 2.0,
+            "b": 2.0,
+            "coord_origin": "TOPLEFT"
+          },
+          "charspan": [
+            0,
+            0
+          ]
+        }
+      ],
+      "captions": [],
+      "references": [],
+      "footnotes": [],
+      "image": null,
+      "data": {
+        "table_cells": [
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 0,
+            "end_row_offset_idx": 1,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "Number of freshwater ducks per year",
+            "column_header": true,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          }
+        ],
+        "num_rows": 1,
+        "num_cols": 1,
+        "grid": [
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 0,
+              "end_row_offset_idx": 1,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "Number of freshwater ducks per year",
+              "column_header": true,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ]
+        ]
+      },
+      "annotations": []
+    },
+    {
+      "self_ref": "#/tables/1",
+      "parent": {
+        "cref": "#/groups/0"
+      },
+      "children": [],
+      "content_layer": "body",
+      "label": "table",
+      "prov": [
+        {
+          "page_no": 1,
+          "bbox": {
+            "l": 1.0,
+            "t": 3.0,
+            "r": 3.0,
+            "b": 10.0,
+            "coord_origin": "TOPLEFT"
+          },
+          "charspan": [
+            0,
+            0
+          ]
+        }
+      ],
+      "captions": [],
+      "references": [],
+      "footnotes": [],
+      "image": null,
+      "data": {
+        "table_cells": [
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 0,
+            "end_row_offset_idx": 1,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "Year",
+            "column_header": true,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 0,
+            "end_row_offset_idx": 1,
+            "start_col_offset_idx": 1,
+            "end_col_offset_idx": 2,
+            "text": "Freshwater Ducks",
+            "column_header": true,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 1,
+            "end_row_offset_idx": 2,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "2019",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 1,
+            "end_row_offset_idx": 2,
+            "start_col_offset_idx": 1,
+            "end_col_offset_idx": 2,
+            "text": "120",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 2,
+            "end_row_offset_idx": 3,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "2020",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 2,
+            "end_row_offset_idx": 3,
+            "start_col_offset_idx": 1,
+            "end_col_offset_idx": 2,
+            "text": "135",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 3,
+            "end_row_offset_idx": 4,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "2021",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 3,
+            "end_row_offset_idx": 4,
+            "start_col_offset_idx": 1,
+            "end_col_offset_idx": 2,
+            "text": "150",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 4,
+            "end_row_offset_idx": 5,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "2022",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 4,
+            "end_row_offset_idx": 5,
+            "start_col_offset_idx": 1,
+            "end_col_offset_idx": 2,
+            "text": "170",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 5,
+            "end_row_offset_idx": 6,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "2023",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 5,
+            "end_row_offset_idx": 6,
+            "start_col_offset_idx": 1,
+            "end_col_offset_idx": 2,
+            "text": "160",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 6,
+            "end_row_offset_idx": 7,
+            "start_col_offset_idx": 0,
+            "end_col_offset_idx": 1,
+            "text": "2024",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          },
+          {
+            "bbox": null,
+            "row_span": 1,
+            "col_span": 1,
+            "start_row_offset_idx": 6,
+            "end_row_offset_idx": 7,
+            "start_col_offset_idx": 1,
+            "end_col_offset_idx": 2,
+            "text": "180",
+            "column_header": false,
+            "row_header": false,
+            "row_section": false,
+            "fillable": false
+          }
+        ],
+        "num_rows": 7,
+        "num_cols": 2,
+        "grid": [
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 0,
+              "end_row_offset_idx": 1,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "Year",
+              "column_header": true,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            },
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 0,
+              "end_row_offset_idx": 1,
+              "start_col_offset_idx": 1,
+              "end_col_offset_idx": 2,
+              "text": "Freshwater Ducks",
+              "column_header": true,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ],
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 1,
+              "end_row_offset_idx": 2,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "2019",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            },
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 1,
+              "end_row_offset_idx": 2,
+              "start_col_offset_idx": 1,
+              "end_col_offset_idx": 2,
+              "text": "120",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ],
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 2,
+              "end_row_offset_idx": 3,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "2020",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            },
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 2,
+              "end_row_offset_idx": 3,
+              "start_col_offset_idx": 1,
+              "end_col_offset_idx": 2,
+              "text": "135",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ],
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 3,
+              "end_row_offset_idx": 4,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "2021",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            },
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 3,
+              "end_row_offset_idx": 4,
+              "start_col_offset_idx": 1,
+              "end_col_offset_idx": 2,
+              "text": "150",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ],
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 4,
+              "end_row_offset_idx": 5,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "2022",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            },
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 4,
+              "end_row_offset_idx": 5,
+              "start_col_offset_idx": 1,
+              "end_col_offset_idx": 2,
+              "text": "170",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ],
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 5,
+              "end_row_offset_idx": 6,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "2023",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            },
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 5,
+              "end_row_offset_idx": 6,
+              "start_col_offset_idx": 1,
+              "end_col_offset_idx": 2,
+              "text": "160",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ],
+          [
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 6,
+              "end_row_offset_idx": 7,
+              "start_col_offset_idx": 0,
+              "end_col_offset_idx": 1,
+              "text": "2024",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            },
+            {
+              "bbox": null,
+              "row_span": 1,
+              "col_span": 1,
+              "start_row_offset_idx": 6,
+              "end_row_offset_idx": 7,
+              "start_col_offset_idx": 1,
+              "end_col_offset_idx": 2,
+              "text": "180",
+              "column_header": false,
+              "row_header": false,
+              "row_section": false,
+              "fillable": false
+            }
+          ]
+        ]
+      },
+      "annotations": []
+    }
+  ],
+  "key_value_items": [],
+  "form_items": [],
+  "pages": {
+    "1": {
+      "size": {
+        "width": 2.0,
+        "height": 7.0
+      },
+      "image": null,
+      "page_no": 1
+    }
+  }
+}
\ No newline at end of file
diff --git a/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.md b/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.md
new file mode 100644
index 00000000..346a85ab
--- /dev/null
+++ b/tests/data/groundtruth/docling_v2/xlsx_05_table_with_title.xlsx.md
@@ -0,0 +1,11 @@
+| Number of freshwater ducks per year   |
+|---------------------------------------|
+
+|   Year |   Freshwater Ducks |
+|--------|--------------------|
+|   2019 |                120 |
+|   2020 |                135 |
+|   2021 |                150 |
+|   2022 |                170 |
+|   2023 |                160 |
+|   2024 |                180 |
\ No newline at end of file
diff --git a/tests/data/xlsx/xlsx_05_table_with_title.xlsx b/tests/data/xlsx/xlsx_05_table_with_title.xlsx
new file mode 100644
index 0000000000000000000000000000000000000000..b7a04de7bc5451d34bca51b7775199d9c659b1cf
GIT binary patch
literal 6335
zcmaJ_1z1!4+urB~fssl|*AP%z1c@;~Qo6y>IR!=wQ%V{n1SCX3L_nmw8zd#9q!}gB
zAm8@B|Lc>#-}k=HuEjY!&+nYy{q(H~#m1oq5D*Xmf_<!v0at_p{oC6~$lSx$$y3P7
z)x*)s)zwkZ+u2DFbLNBgJ%?5>McpdMYUeX5FI1a<m6E7Be;qrtu0qWTVt2>+$62Ae
z-F1R&{H(S<2iP;q-ZA&Jb}prjIvq_L2?BV>M1~{C4>g-<W}(;LBXF2DLR)6a68Gdf
zPf6c`p6V?ceDzfFwc)%y_n}u7PPRjjJ{l+POPlwIj|Xii={Pyyy6e!gJb4>*`Q=c3
zgx<v0lWu75264ANZNlqlfwHIO-kg#A;lXuTma?{Zi#joIQumG+6s^uw;7Ju%WY+{v
zpf1tTzBFplXIQW(+Mv->ty8SnM$NpXPu0iMW9*J56rZ4BNRoaYjY0ty0HFDgIFbL2
z(^Z6C=uCU0gDL!X^udTq;GBamKg{GhXR3X=#w*T2d48qOv2>8*VU5fBNU*)+mI5<i
zOA$6WY5MTasw$o`_@SL5L^?_Z5V2|UPKLugYq<Fp&bV-rNC8OG0|<2)-+9R8OqM$S
z9I(R&g{+&@lPL5;-=*BxV4zkaa^_tb5<MFl$1`w&0`$Axs1-kzSjoqcS=#@?SD;1C
zmyFdF$W@cI1=)~IPHT@Q7wIEocV`HR3`_$bfrd3wmTsCf4*#-x8nvgO^EO2t(wv7&
zZL~3>a7bFEps1!wxzu;BE{xN^hnZr^k~8@pY}1>@t*r6;hxKba<=5sud#G(!ebLrl
zRP+4s?0&}lWdv$UNv5>#a9}@DvszZ`DSJ(wMArG9;!<%NaG+x!vQvv)Ds+Qm-&>e1
z*m}}>np-vhni*ec=>Y8>H%ZcU;O1u<BBD9&A0nHL!9Pg9-9D#~wxQ`O-3;huXg!jM
z-{t!b-sCRox*2~58VvNzijnb-3G0!FmKO)7WUVQ&rdPdzy*|$K4E@S7CH715?fZJ%
zb-Eunl(eWTtK6P2f7qYfd<c25+ISlp(I;044fvokM0c+{=IGNG{?tLGkT0A}<cr5o
zB#*(FYWVBUX;xGT&SffrixEDf9I}wkImwXAUUuqQ;^-5#?yc028E*~N{Ku{j1W*&!
z7sMC`O3-Us#z8xeFZRDV5Y^uf1h=zwwiWv8Qy3Fw{rb%w(<~JAM|^bKgz@|mWn53l
zATl!Wd=d3iAOUMQ9xLfk-mf#Z5)NvK;t|P=(;!QqI#VW;mHFIxk%V`ScBxX4g*UyN
z9Im|wXAkK7u%udrRg)p|`o7W-Xhs{?b)30)Pg}n{zlopLAzZ&?xW<$%FHG3_@npGf
zN_9@0*!UxEXd?97$<8r4hb1MKK?LZvxy-Yfo?{U@E4HMpj(l=6-lIr;t<dVjQ@Q!6
zw&QlJW(8UnlO2VNF8fN2$9OHvGNpTGeEhcqNos{SdQr*H-Ov*vyhH3Srwj*x;<r^T
z0Z*!KumnbNR+WU&>iZ1RN;L39YpdFQ-8TeqwxO&xqy&e-)1C;Hi42bf7Q#j0+3#8E
zyOI38@$Ce68F%+^vheoll-d2H=M(JDQ&Ur8cDsYJ9(-;v>iaG{`;>Q}?Rzs7>OuZ<
zYSK$MMJ9uH&D~ufvM&AmIPt}lOyS;<k_$-_Q*m`3)|xYvSdC<xW+K&w?rE~UwzFIR
z@i-14u%JX&H=S=-q7xcXuQzail$Z*!Q1oaq1vU`1x$ndDiq)Kg@4*&1-RcpdcO7Pn
ztwRbs>7d(cYC)V#_rSAlQLvdVb6r*GpJUM#y04veD|6ZF^4Y-I83KFSzNjKy*@YtP
zOjXHR$rqw<<nsM-utZ0MG^p<}Zq@tw1!)h#Ahwi|y1Qom;uosncr%#|6(7o-$3GAB
zsw_6f8taL_m8x$Y$%6!bjOwgO$#~$Fbz3%Mu{nz;I$xr--5V!UJq(hVf@=T36V5rQ
zUg`2eheMSBEYzQ?Q9iBDm(<W&=wbY+pghp1t%;Vyo%Z~M0A}yaB>z(zU_`AnsL$Uz
z)F%-i@@W-6S0T=JdF8G^&|Ww$H~?A@arDOFl}4mj_wd>p**U*+X|dcyid5&#h3$gv
zVN~|yYQ8T;&3T=h*gNQO#+RE8DTrCg!a#F%{L~Su!R;4M2WBBdOF^04^h@_g&gFBn
znC*TrTeuotVqt<tDtnITCpu`bNdGNp(0^j8lPBEA$<`ASE?Wks52nG?{&OEo&Sm41
zH7!&_=ycxJtJzT#+E0a~i-#kvl`Duz8v`cX(>QStwL%}}OLNFJtj;8C#?+7vKEze_
z(zWMor;Z`>i+UGk+j9SQ`^5g_8C6s{D;ui`kz_k)?3f=x>!9O~NYZhj;3kl1uNGd$
z9~^PlAkHDTf#Ckk(+0K4>DXopf;hq}@}Bxy&t-YzT4E_@8_;}ev7_#N1GdOT#`^j}
z0IzW9B&vy1gOUbyH5J7VbA}voa==Mext^gnD3GVXZ`a7!*cq9if}mR#iBmQ~1lpRk
z?)SuvxgRwrQb*~NF-_FH;wCjm<!!WS1n`B=y_DAPn?w<v`YM@Mk>w_j9txQ1v`Wv5
z?e>6*n4g@}t!^E^D8<exx!Iw~@n}e{|E!LewKB+X-Daa|?mjF+H+N6#x#nG#9r;3>
zhxfejBndd8Y_ur?Of}q9tMaRmvfaL=c)777qd_HBhAvr3$(bNwC7Ru`m9L{3)&9Xh
zg_qBJ1PjvP`v<aN#lxLYxlvMsf^pydjl?uO9;jJ=9;l#m1jxJ@%LE@V<Rq#Ac^Bkm
zmlk-1GDORlk=?J>MFI1+0FU~K${E8fOOZ22RyJ!rehJUSLA&seDiZi50&G-}47yWz
zaY|K2FTG=JwwkRYwmbdYu#NLA7@?+A+$`_6!o<dc5xWUVG1*GuF=TC$(Swjm>qpWl
zJ{^4nj*eiF6#71rM|dK&DTHK4>o0H7Z6&2!e4L3o>h52^d;9z2n4N9#oJ1#nEz@`X
zb?a5ryF%r^#s_}Zm*EELSKMXZz1$2Og62<Ye$qG1bZxLR`!x92MQBRK^{EthnU{jj
z`Pl5YJ&kWP&<;|n;$<05ldc?=&EqdxF>c5FK>wq_@;pe4s84*kZbU?tgG8r{#x9?R
z&;%sL>zsNgzL2it^_y3R+}majI%;-K#fG_Am^d(kJR$2x$H9Bhzr_LF@9x9X&ej(0
zDfHLZUlpWYe>!^lHl__BK*VtL93i(@_TJc+7K9*10RFigJ#5Krdhr!rN1IZ4g30z$
z220JZiwkxuujq*+U-|iUr+TqyYc;=>-3V<^_U5PT@zS|G8`)^>I!f%2ZdP-<bzmO;
z<L*E}`gNq=LweS!5F8Wr3$@KfRs!8uao68Ra}zwz(CxbJXwYN&nh4@%W!_^*fm~qe
z5?&5SUAWi4^;|`VApse7B(YGV#zxr6u=YNRrf}dezK+kFZSPv%xStU<?XpG!`GMe%
zWzl5$a9uB}IQ>yUz37m=)B#%h3nR`^`~rLjPS9BQk7F5LPyVc<cKj|=xwjRGS{Dwu
zXP?6RFVzt|wrO1-*@9W$rto0-;?BmAewib#{g~Bu2e7eaahtI4B`hJdMVdt9K|FjT
zP>CE$okqd9N6w)<hB(yajOJshf*FjR@E_Oab+a$yXbK#Gjw{6naL=&%gFL5Zk6+47
zILHl~;!!h03eyuAT<?U&86@Og%pyM0Q%}~d7&oWk5JL7c?G=DMgofYoEu;^s*SSIx
zbHE%cp``_vgePWBR+mBNA(=b%gmJD`eXpV1B*vSKs7*nK(~~gs_SuRzQ({@;55r*C
zsdd?@@p_i0#uoeZ5BdGbu@NjB>36@BkS6XRHttGMewvKx%X7Iq(P}uUfIU)-0Qw)_
zqsmaf(1ML_#niS2jBus&&{`qL%7OkNhNGK>XE{#$>=G1sUy{`sLr`yO=8L@BZG*)!
z)$_y=V8W5r`e8PqkTlB22T`1a3#29;ze1yzu@bDSG*_HPxnys*%PNT3I^co8Lc#BV
zG=e4k!NtzPZZ2WUUK~t%Z8F}5Ktu{MVIGjUXarf}bq}_MD?pBlRcpvTR1@=?SWMR6
ztMk%dtdAJC$LXSPw!IH$DqX7{M%>9e*x~&!$2I_6WMz=r5Dd{=y56EVH#UJa@RM}A
z(A7;btJgn(KD5SPd9G>fPF1$6x4wU?<LS(Z4}IwA_W85oh6Pgp+}aIB(~(s_<NaP1
zQTV|N6aQSs(+;n(I}Q0CH`i)O<gxYVzM_w^$&p7FJI^Ml53eKjbB4N-zT#y38p`Io
zWp@%KuU$NWtXKA?H&EaG#pPHdHK|Pt3^ZYES5rCVtm<$gdr8Ii0IQ!AI&$p6&aHl^
z&+kP|B;sh+8eM8w=44e3XoOvE8fRxW9}0g8_=@xM8UrhivL%J0C3!=;E`h&4pPkx8
z1AxzvqB`5Pjmc)L?87)pe}G)Cw*z&O2`NpkK(Ixh^RWAo5?PMMKb7-kOg}}&8d(u>
zKJIlpKM&`+d#f=0ECk={W3paMbLoJhOY!5LD(8UlA;o(Om#}wZ8Pcq!+}s5hMw3ZN
z&OtASM3F9@^*P!Zs$#p6kmTnhdl~BY;}a)H#IJvVdGm`TyY0^i;`Od*NS`rdvJbn>
z{DwFV0HBW^Z~l4qq5e-5wzKrGwb6ll*t@{~D#kBjg&$xGlOVkU6qhnqxGClJ;>mPx
zq*_>zNSF(4v=f%<tTl2e%73Fs&&rxRyP3YFZ!}8C*QrLc{SccwFSb(!pZKHD*|f4I
zf2LBSjXW{T=4%aOD}SMKxIiD9ZLZjJ`h-<IL6vW*9_nT_Mfm-1a+rOT63f>Pz|w0=
z?7>9UaRycRW849vey3vtZtxZS(uG9@t|PJzguL6JK)tn|kG9TSf0kJdgRyuSufJsO
zN2N%S)8Rek!RbKqBmKVZr`6Bkmmz0(7%1eo>vy`*8&w4U8x+LoEB>C9|DKQIn>AX&
zB+qtyBY2O*zR3|}=hv#K)Ut{)HQjV!{2ZR!7V_ipJ{2X&i?Y^e_kh4(R&Qq2T=-|f
zoKUlHJ;QLUdQNNqw{5+@+<k;@_vZw3ToYH;c2(f@h<Tv;vv1*EPo6J4R+VUNUmf}d
zxN42E0m#Tb_b}D-y7kK@Y$enX6-DU|dzkYz5T`Cw-Puvh1;TtCAuF<|&z8XYq)SyT
z*YB-(9VMylGa{c^TmS0bxt&BZdJ?dT`n8v=E8AT(BbCvt7m2;=DJfJn)hixuRzLCj
z>nKCPFu1QeyT6Lsb{6P{h`DN{QRaE?3Ka$>WSVH&HX4Khbj|u_kGY1v!p7BF)5F!x
zQ^?xY!xq!KA+>v5TF_*|5g*$e6<07DHIn6Xn2r`7&|xgS^yPfKc=Fcn@tGtUp(d-|
z4@)nYbnW-Yk}^q1FTfig4o+?&`K933pzmy1uMZryW8&VDUW=km3L=yYQ5r$<yt$Q~
z^pso=qWLW3uqbwjFo~*MM2VBmu_h#bt?5=MZe<d0cxgdT=}q_pk0y7zXg9)VU@HO_
z9REqaoXDHG^?5~l3pjM`nsS6KcGq#QHF%J|)4p3bAItctyWg-y!2$mem9QeS=Pixb
z8Js(H!hZhQ+gS3wEN#DzmYgAi{$@5&7;wcfIL~>4H7jId(jmG`JSyvgyBr3JPOx+u
z^oqEgamAUg(l!h+qhb^p(sj2-ks0q6TkJm#tE(W-4Bnl9Dz8KPd5LrSTLk@kgX>>?
z=m;^oqvca@;mGr&)f@C8o?D~E7@85ZSv232-uRN#bVSxUfBawt|DC<SLHG;OJGeGk
zbr!Ql`Mx{;x%Y0USnEr9K;0sa8b!}=F^1!%;_E=5eS7a;4M%a+aF%Xv7`tV}syqJW
z<Y%DfpK0QxhDH(J%UnqCZZNV7)Zz|mF~l5WPxGudW=#*r?8D>D!j*@i$+@Up83V3z
z-Y=pMEu4r=rFW*`x5pf%mE-o+#EK?}RN;W<$v=brjGUVgKtUp$Wz|t1R~kC0*<c$%
zLT8ZVhl=<U&8ZvX&V1r^cCsp(OfY*R3p$y1cxhivg4fhcP#1yOFIGCwW-268nsW27
z>=7%$lzvdYEkkocLO9Km<GswNbzGRNxctOoXG^C&?!b?m&_|HlhQwk~2S;`9BBg86
z-?frAHpvWpf|oJepKGIUKB{8(^}U_0!JHu4W<Q@32NNLG!9T4)#56PkKh?1x=TQ26
z%J>DufXO!B=|heNv-V$;1!m;`9R-g);jYeq;W<!B)B>hPE_{1SFgdM@A2(bcDK4dZ
z6gyi1^mOe~kM#Ftf4Uw0oLN?5MCL-Fh+^`ZL$wee4?uvFj7z}v;*B_@xqIG%DfIO+
z9)InV2@`hQux?-^KZ~QpTWW_lUkBz%UrJaDvMai3XS!>PMh~>4q85P(lDcDBlSrx)
zHb04zy$!JoD4EUN&he}(tFLoz0Bo8Tv#*75K+sF{Q}3bQ{~X<zeHkmby1;E+;O4qM
zZnmDLn0?jjb)%CZhRR?w*IG=*mgjaR=U0ArUxMBG0p*By+*b184n3D<(lC-2^wMCj
z&-l@;()~T{v2LtaFO|r<SvC`JMDNINc&M1=ldp@>#~@DNw#{5A?=NC?S-&!>%6ya)
z&Y~w~U`M(~;QBr0_#8E8K=`w4t}$n34Ir7v>Jh6IfRCF#!CGy;_~d=0Tl`FgkCuZ<
z%ID8Z{1z!4+~2}IHP5h{kTk~I!$L56w)E@fkWHaOJ_A!1LKP5S<XP>t9hxT!X#lIt
zl5ZMBDx7^t<$Q3kpyMe+WHxh0|7=vMtsRe)r;l&w9gYV&B8sdM1g)z?tf`e`DZ3y!
zx&8h$9?pYn7eUrITb@Ic3JRY0ho0+^{+e07ofl~L7<;$eux8?Hq%G~z0H1e7P8$Dt
z<H;mrX}5ouvefRS@X{K5G3vmZT0*>j#<f07Lx3`v_{VOg86Vr@)nD?`COBa_pHF7+
zb0vE#MH}0FC7g?qU$`(f<4(m(@kINfA|A$OKrCv&Z>IMuw~1kT|6Bf<_x*F7s}vW8
z3H>eH=sTm=xyp<FseE<R$GFaKu}3pzXyrdSu|F5MIwN5if!~se-t&L)1AnSsZE!F{
z)Njcr{ND%vKO@zjt6c33FwM|!xk>i-1N_+<{kg)`vWe*-ev1$q&nq3ppNd!0Hm0in
zmWSwE{X4t-r^f!Nel<~Ha{F%~r@W&6pG^O!_SJOo-<C5$AO61i|E9J-7r5%1|Gj`H
d8pr=9hiF1^(Iy1|h|#ZRbfsEDkHrAM{{Wx=ih=+D

literal 0
HcmV?d00001

diff --git a/tests/test_backend_msexcel.py b/tests/test_backend_msexcel.py
index 6084a4b6..31cb59b4 100644
--- a/tests/test_backend_msexcel.py
+++ b/tests/test_backend_msexcel.py
@@ -6,9 +6,10 @@ import pytest
 from openpyxl import load_workbook
 
 from docling.backend.msexcel_backend import MsExcelDocumentBackend
+from docling.datamodel.backend_options import MsExcelBackendOptions
 from docling.datamodel.base_models import InputFormat
 from docling.datamodel.document import ConversionResult, DoclingDocument, InputDocument
-from docling.document_converter import DocumentConverter
+from docling.document_converter import DocumentConverter, ExcelFormatOption
 
 from .test_data_gen_flag import GEN_TEST_DATA
 from .verify_utils import verify_document, verify_export
@@ -227,6 +228,49 @@ def test_inflated_rows_handling(documents) -> None:
     )
 
 
+def test_table_with_title():
+    """Test that singleton cells with non-numeric content are treated as TextItem.
+
+    When treat_singleton_as_text option is enabled, 1x1 tables containing non-numeric
+    text should be converted to TextItem instead of TableItem. This test verifies that
+    xlsx_05_table_with_title.xlsx is correctly parsed with this option.
+    """
+    path = next(
+        item for item in get_excel_paths() if item.stem == "xlsx_05_table_with_title"
+    )
+
+    # Create converter with treat_singleton_as_text=True
+    options = MsExcelBackendOptions(treat_singleton_as_text=True)
+    format_options = {InputFormat.XLSX: ExcelFormatOption(backend_options=options)}
+    converter = DocumentConverter(
+        allowed_formats=[InputFormat.XLSX], format_options=format_options
+    )
+
+    conv_result: ConversionResult = converter.convert(path)
+    doc: DoclingDocument = conv_result.document
+
+    # With treat_singleton_as_text=True, the singleton title cell should be a TextItem
+    texts = list(doc.texts)
+    tables = list(doc.tables)
+
+    assert len(texts) == 1, f"Should have 1 text item (the title), got {len(texts)}"
+    assert len(tables) == 1, f"Should have 1 table, got {len(tables)}"
+
+    # Verify the text item contains the title
+    assert texts[0].text == "Number of freshwater ducks per year", (
+        f"Text should be 'Number of freshwater ducks per year', got '{texts[0].text}'"
+    )
+
+    # Verify table dimensions
+    table = tables[0]
+    assert table.data.num_rows == 7, (
+        f"Table should have 7 rows, got {table.data.num_rows}"
+    )
+    assert table.data.num_cols == 2, (
+        f"Table should have 2 columns, got {table.data.num_cols}"
+    )
+
+
 def test_bytesio_stream():
     """Test that Excel files can be loaded from BytesIO streams.