docling-core 2.2.1__py3-none-any.whl → 2.2.3__py3-none-any.whl

docling_core/transforms/chunker/hierarchical_chunker.py

@@ -129,8 +129,8 @@ class HierarchicalChunker(BaseChunker):
         table_df.index = table_df.index + 1
         table_df = table_df.sort_index()
 
-        rows = [item.strip() for item in table_df.iloc[:, 0].to_list()]
-        cols = [item.strip() for item in table_df.iloc[0, :].to_list()]
+        rows = [str(item).strip() for item in table_df.iloc[:, 0].to_list()]
+        cols = [str(item).strip() for item in table_df.iloc[0, :].to_list()]
 
         nrows = table_df.shape[0]
         ncols = table_df.shape[1]

docling_core/types/doc/document.py

@@ -4,6 +4,7 @@ import base64
 import mimetypes
 import re
 import sys
+import textwrap
 import typing
 from io import BytesIO
 from typing import Any, Dict, Final, List, Literal, Optional, Tuple, Union
@@ -1125,6 +1126,7 @@ class DoclingDocument(BaseModel):
         image_placeholder: str = "<!-- image -->",
         image_mode: ImageRefMode = ImageRefMode.PLACEHOLDER,
         indent: int = 4,
+        text_width: int = -1,
     ) -> str:
         r"""Serialize to Markdown.
 
@@ -1207,8 +1209,8 @@ class DoclingDocument(BaseModel):
             elif isinstance(item, TextItem) and item.label in [DocItemLabel.TITLE]:
                 in_list = False
                 marker = "" if strict_text else "#"
-                text = f"{marker} {item.text}\n"
-                mdtexts.append(text.strip())
+                text = f"{marker} {item.text}"
+                mdtexts.append(text.strip() + "\n")
 
             elif (
                 isinstance(item, TextItem)
@@ -1251,7 +1253,10 @@ class DoclingDocument(BaseModel):
 
             elif isinstance(item, TextItem) and item.label in labels:
                 in_list = False
-                if len(item.text):
+                if len(item.text) and text_width > 0:
+                    wrapped_text = textwrap.fill(text, width=text_width)
+                    mdtexts.append(wrapped_text + "\n")
+                elif len(item.text):
                     text = f"{item.text}\n"
                     mdtexts.append(text)
 

docling_core/types/doc/labels.py

@@ -29,6 +29,10 @@ class DocItemLabel(str, Enum):
     PARAGRAPH = "paragraph"  # explicitly a paragraph and not arbitrary text
     REFERENCE = "reference"
 
+    def __str__(self):
+        """Get string value."""
+        return str(self.value)
+
 
 class GroupLabel(str, Enum):
     """GroupLabel."""
@@ -43,6 +47,10 @@ class GroupLabel(str, Enum):
     SHEET = "sheet"
     SLIDE = "slide"
 
+    def __str__(self):
+        """Get string value."""
+        return str(self.value)
+
 
 class TableCellLabel(str, Enum):
     """TableCellLabel."""
@@ -51,3 +59,7 @@ class TableCellLabel(str, Enum):
     ROW_HEADER = "row_header"
     ROW_SECTION = "row_section"
     BODY = "body"
+
+    def __str__(self):
+        """Get string value."""
+        return str(self.value)

{docling_core-2.2.1.dist-info → docling_core-2.2.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling-core
-Version: 2.2.1
+Version: 2.2.3
 Summary: A python library to define and validate data types in Docling.
 Home-page: https://ds4sd.github.io/
 License: MIT

{docling_core-2.2.1.dist-info → docling_core-2.2.3.dist-info}/RECORD

@@ -16,13 +16,13 @@ docling_core/search/package.py,sha256=Lz2ml2eDy5t0ZimnGTq-DXHAn-f18w0bn4H5xrhs75
 docling_core/transforms/__init__.py,sha256=P81y_oqkiTN4Ld5crh1gQ6BbHqqR6C6nBt9ACDd57ds,106
 docling_core/transforms/chunker/__init__.py,sha256=sSSTnt7ZCt8Og1e0jhApNTtA0pyyHyzwcl8yXFLb2J8,292
 docling_core/transforms/chunker/base.py,sha256=iPouZOJ3cYWvai4P0Gpd3QmsTKQuY5fFUXzTMk_XNmE,1571
-docling_core/transforms/chunker/hierarchical_chunker.py,sha256=-wZrQAPMiROb9yKiSN3eGqtD13tjcDVAiDUz971Q8PI,8087
+docling_core/transforms/chunker/hierarchical_chunker.py,sha256=uG7nNoUCFqWeQAKydQg731JYJ9sayUe4J48nMF0VHE8,8097
 docling_core/types/__init__.py,sha256=MVRSgsk5focwGyAplh_TRR3dEecIXpd98g_u3zZ5HXo,260
 docling_core/types/base.py,sha256=PusJskRVL19y-hq0BgXr5e8--QEqSqLnFNJ8UbOqW88,8318
 docling_core/types/doc/__init__.py,sha256=bEL4zKVOG7Wxm6xQrgF58mu-Teds9aSavuEAKVNhrTU,639
 docling_core/types/doc/base.py,sha256=zvx631U_yQCcJam83hNdDanXEYnO3eN-CCw9vDr6S-I,4442
-docling_core/types/doc/document.py,sha256=SrOXpO6iCIYFkhWW-pksd4C4PeJ2jubKI5m34K_lTac,51902
-docling_core/types/doc/labels.py,sha256=mzmSd072A-qW3IThswHxwIHV8IoyTCbHHlNOrisinRA,1335
+docling_core/types/doc/document.py,sha256=B56FA5lGAEodjfIUncXSstQclAmyt3GOybMiKEEIc7s,52138
+docling_core/types/doc/labels.py,sha256=A8vWP82VAeXO1rlCO0oDKo_Hb8uDeQe0myOTY3P03hk,1596
 docling_core/types/gen/__init__.py,sha256=C6TuCfvpSnSL5XDOFMcYHUY2-i08vvfOGRcdu6Af0pI,124
 docling_core/types/gen/generic.py,sha256=l4CZ4_Lb8ONG36WNJWbKX5hGKvTh_yU-hXp5hsm7uVU,844
 docling_core/types/legacy_doc/__init__.py,sha256=Pzj_8rft6SJTVTCHgXRwHtuZjL6LK_6dcBWjikL9biY,125
@@ -49,8 +49,8 @@ docling_core/utils/generate_docs.py,sha256=BdKAoduWXOc7YMvcmlhjoJOFlUxij1ybxglj6
 docling_core/utils/generate_jsonschema.py,sha256=uNX1O5XnjyB5nA66XqZXTt3YbGuR2tyi_OhHepHYtZg,1654
 docling_core/utils/validate.py,sha256=3FmnxnKTDZC5J9OGxCL3U3DGRl0t0bBV1NcySXswdas,2031
 docling_core/utils/validators.py,sha256=azcrndLzhNkTWnbFSu9shJ5D3j_znnLrIFA5R8hzmGU,2798
-docling_core-2.2.1.dist-info/LICENSE,sha256=2M9-6EoQ1sxFztTOkXGAtwUDJvnWaAHdB9BYWVwGkIw,1087
-docling_core-2.2.1.dist-info/METADATA,sha256=k6jXQiT3A4us_DXcTvnIi2Bk1UMF6GiJ23Zj9MLbcNg,5432
-docling_core-2.2.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling_core-2.2.1.dist-info/entry_points.txt,sha256=jIxlWv3tnO04irlZc0zfhqJIgz1bg9Hha4AkaLWSdUA,177
-docling_core-2.2.1.dist-info/RECORD,,
+docling_core-2.2.3.dist-info/LICENSE,sha256=2M9-6EoQ1sxFztTOkXGAtwUDJvnWaAHdB9BYWVwGkIw,1087
+docling_core-2.2.3.dist-info/METADATA,sha256=DlV-TrYKPq-qbI9d0iS4mrOJs_CwV9QZNflqGEy0crE,5432
+docling_core-2.2.3.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling_core-2.2.3.dist-info/entry_points.txt,sha256=jIxlWv3tnO04irlZc0zfhqJIgz1bg9Hha4AkaLWSdUA,177
+docling_core-2.2.3.dist-info/RECORD,,