docling-core 2.12.1__tar.gz → 2.13.0__tar.gz

{docling_core-2.12.1 → docling_core-2.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling-core
-Version: 2.12.1
+Version: 2.13.0
 Summary: A python library to define and validate data types in Docling.
 Home-page: https://ds4sd.github.io/
 License: MIT

{docling_core-2.12.1 → docling_core-2.13.0}/docling_core/types/doc/document.py

@@ -14,7 +14,7 @@ import warnings
 from io import BytesIO
 from pathlib import Path
 from typing import Any, Dict, Final, List, Literal, Optional, Tuple, Union
-from urllib.parse import unquote
+from urllib.parse import quote, unquote
 
 import pandas as pd
 import yaml
@@ -830,7 +830,7 @@ class PictureItem(FloatingItem):
             ):
                 return default_response
 
-            text = f"\n![Image]({str(self.image.uri)})\n"
+            text = f"\n![Image]({quote(str(self.image.uri))})\n"
             return text
 
         else:
@@ -884,7 +884,7 @@ class PictureItem(FloatingItem):
             ):
                 return default_response
 
-            img_text = f'<img src="{str(self.image.uri)}">'
+            img_text = f'<img src="{quote(str(self.image.uri))}">'
             return f"<figure>{caption_text}{img_text}</figure>"
 
         else:

{docling_core-2.12.1 → docling_core-2.13.0}/docling_core/types/doc/labels.py

@@ -1,12 +1,12 @@
 """Models for the labels types."""
 
 from enum import Enum
+from typing import Tuple
 
 
 class DocItemLabel(str, Enum):
     """DocItemLabel."""
 
-    # DocLayNet v2
     CAPTION = "caption"
     FOOTNOTE = "footnote"
     FORMULA = "formula"
@@ -26,12 +26,34 @@ class DocItemLabel(str, Enum):
     KEY_VALUE_REGION = "key_value_region"
 
     # Additional labels for markup-based formats (e.g. HTML, Word)
-    PARAGRAPH = "paragraph"  # explicitly a paragraph and not arbitrary text
+    PARAGRAPH = "paragraph"
     REFERENCE = "reference"
 
-    def __str__(self):
-        """Get string value."""
-        return str(self.value)
+    @staticmethod
+    def get_color(label: "DocItemLabel") -> Tuple[int, int, int]:
+        """Return the RGB color associated with a given label."""
+        color_map = {
+            DocItemLabel.CAPTION: (255, 204, 153),
+            DocItemLabel.FOOTNOTE: (200, 200, 255),
+            DocItemLabel.FORMULA: (192, 192, 192),
+            DocItemLabel.LIST_ITEM: (153, 153, 255),
+            DocItemLabel.PAGE_FOOTER: (204, 255, 204),
+            DocItemLabel.PAGE_HEADER: (204, 255, 204),
+            DocItemLabel.PICTURE: (255, 204, 164),
+            DocItemLabel.SECTION_HEADER: (255, 153, 153),
+            DocItemLabel.TABLE: (255, 204, 204),
+            DocItemLabel.TEXT: (255, 255, 153),
+            DocItemLabel.TITLE: (255, 153, 153),
+            DocItemLabel.DOCUMENT_INDEX: (220, 220, 220),
+            DocItemLabel.CODE: (125, 125, 125),
+            DocItemLabel.CHECKBOX_SELECTED: (255, 182, 193),
+            DocItemLabel.CHECKBOX_UNSELECTED: (255, 182, 193),
+            DocItemLabel.FORM: (200, 255, 255),
+            DocItemLabel.KEY_VALUE_REGION: (183, 65, 14),
+            DocItemLabel.PARAGRAPH: (255, 255, 153),
+            DocItemLabel.REFERENCE: (176, 224, 230),
+        }
+        return color_map[label]
 
 
 class GroupLabel(str, Enum):

{docling_core-2.12.1 → docling_core-2.13.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "docling-core"
-version = "2.12.1"
+version = "2.13.0"
 description = "A python library to define and validate data types in Docling."
 license = "MIT"
 authors = [