docling-core 2.48.0__tar.gz → 2.48.2__tar.gz

{docling_core-2.48.0 → docling_core-2.48.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling-core
-Version: 2.48.0
+Version: 2.48.2
 Summary: A python library to define and validate data types in Docling.
 Author-email: Cesar Berrospi Ramis <ceb@zurich.ibm.com>, Panos Vagenas <pva@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Christoph Auer <cau@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>
 Maintainer-email: Panos Vagenas <pva@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Christoph Auer <cau@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>, Cesar Berrospi Ramis <ceb@zurich.ibm.com>

{docling_core-2.48.0 → docling_core-2.48.2}/docling_core/transforms/serializer/markdown.py

@@ -332,7 +332,7 @@ class MarkdownTableSerializer(BaseTableSerializer):
                 ]
                 for row in item.data.grid
             ]
-            if len(rows) > 1 and len(rows[0]) > 0:
+            if len(rows) > 0:
                 try:
                     table_text = tabulate(rows[1:], headers=rows[0], tablefmt="github")
                 except ValueError:

{docling_core-2.48.0 → docling_core-2.48.2}/docling_core/types/doc/document.py

@@ -4386,6 +4386,7 @@ class DoclingDocument(BaseModel):
         to_element: int = sys.maxsize,
         labels: Optional[set[DocItemLabel]] = None,
         strict_text: bool = False,
+        escape_html: bool = True,
         escaping_underscores: bool = True,
         image_placeholder: str = "<!-- image -->",
         image_mode: ImageRefMode = ImageRefMode.PLACEHOLDER,
@@ -4414,6 +4415,7 @@ class DoclingDocument(BaseModel):
             to_element=to_element,
             labels=labels,
             strict_text=strict_text,
+            escape_html=escape_html,
             escape_underscores=escaping_underscores,
             image_placeholder=image_placeholder,
             image_mode=image_mode,
@@ -4435,6 +4437,7 @@ class DoclingDocument(BaseModel):
         to_element: int = sys.maxsize,
         labels: Optional[set[DocItemLabel]] = None,
         strict_text: bool = False,
+        escape_html: bool = True,
         escape_underscores: bool = True,
         image_placeholder: str = "<!-- image -->",
         enable_chart_tables: bool = True,
@@ -4465,6 +4468,8 @@ class DoclingDocument(BaseModel):
         :type labels: Optional[set[DocItemLabel]] = None
         :param strict_text: Deprecated.
         :type strict_text: bool = False
+        :param escape_html: bool: Whether to escape HTML reserved characters in the
+            text content of the document. (Default value = True).
         :param escape_underscores: bool: Whether to escape underscores in the
             text content of the document. (Default value = True).
         :type escape_underscores: bool = True
@@ -4511,6 +4516,7 @@ class DoclingDocument(BaseModel):
                 pages={page_no} if page_no is not None else None,
                 start_idx=from_element,
                 stop_idx=to_element,
+                escape_html=escape_html,
                 escape_underscores=escape_underscores,
                 image_placeholder=image_placeholder,
                 enable_chart_tables=enable_chart_tables,

{docling_core-2.48.0 → docling_core-2.48.2}/docling_core.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling-core
-Version: 2.48.0
+Version: 2.48.2
 Summary: A python library to define and validate data types in Docling.
 Author-email: Cesar Berrospi Ramis <ceb@zurich.ibm.com>, Panos Vagenas <pva@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Christoph Auer <cau@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>
 Maintainer-email: Panos Vagenas <pva@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Christoph Auer <cau@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>, Cesar Berrospi Ramis <ceb@zurich.ibm.com>

{docling_core-2.48.0 → docling_core-2.48.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "docling-core"
-version = "2.48.0"  # DO NOT EDIT, updated automatically
+version = "2.48.2"  # DO NOT EDIT, updated automatically
 description = "A python library to define and validate data types in Docling."
 license = "MIT"
 license-files = ["LICENSE"]

{docling_core-2.48.0 → docling_core-2.48.2}/test/test_serialization.py

@@ -25,7 +25,13 @@ from docling_core.transforms.serializer.markdown import (
 )
 from docling_core.transforms.visualizer.layout_visualizer import LayoutVisualizer
 from docling_core.types.doc.base import ImageRefMode
-from docling_core.types.doc.document import DoclingDocument, MiscAnnotation, TableItem
+from docling_core.types.doc.document import (
+    DoclingDocument,
+    MiscAnnotation,
+    TableCell,
+    TableData,
+    TableItem,
+)
 from docling_core.types.doc.labels import DocItemLabel
 
 from .test_data_gen_flag import GEN_TEST_DATA
@@ -317,6 +323,29 @@ def test_md_rich_table():
     verify(exp_file=exp_file, actual=actual)
 
 
+def test_md_single_row_table():
+    exp_file = Path("./test/data/doc/single_row_table.gt.md")
+    words = ["foo", "bar"]
+    doc = DoclingDocument(name="")
+    row_idx = 0
+    table = doc.add_table(data=TableData(num_rows=1, num_cols=len(words)))
+    for col_idx, word in enumerate(words):
+        doc.add_table_cell(
+            table_item=table,
+            cell=TableCell(
+                start_row_offset_idx=row_idx,
+                end_row_offset_idx=row_idx + 1,
+                start_col_offset_idx=col_idx,
+                end_col_offset_idx=col_idx + 1,
+                text=word,
+            ),
+        )
+
+    ser = MarkdownDocSerializer(doc=doc)
+    actual = ser.serialize().text
+    verify(exp_file=exp_file, actual=actual)
+
+
 # ===============================
 # HTML tests
 # ===============================