docling-core 2.2.1__tar.gz → 2.2.3__tar.gz

{docling_core-2.2.1 → docling_core-2.2.3}/PKG-INFO

@@ -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 → docling_core-2.2.3}/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-2.2.1 → docling_core-2.2.3}/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-2.2.1 → docling_core-2.2.3}/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 → docling_core-2.2.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "docling-core"
-version = "2.2.1"
+version = "2.2.3"
 description = "A python library to define and validate data types in Docling."
 license = "MIT"
 authors = [