docling-core 2.33.0__tar.gz → 2.33.1__tar.gz

{docling_core-2.33.0 → docling_core-2.33.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling-core
-Version: 2.33.0
+Version: 2.33.1
 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>
@@ -32,7 +32,7 @@ Requires-Dist: pandas<3.0.0,>=2.1.4
 Requires-Dist: pillow<12.0.0,>=10.0.0
 Requires-Dist: pyyaml<7.0.0,>=5.1
 Requires-Dist: typing-extensions<5.0.0,>=4.12.2
-Requires-Dist: typer<0.16.0,>=0.12.5
+Requires-Dist: typer<0.17.0,>=0.12.5
 Requires-Dist: latex2mathml<4.0.0,>=3.77.0
 Provides-Extra: chunking
 Requires-Dist: semchunk<3.0.0,>=2.2.0; extra == "chunking"

{docling_core-2.33.0 → docling_core-2.33.1}/docling_core/types/doc/document.py

@@ -3237,6 +3237,11 @@ class DoclingDocument(BaseModel):
             "document_index": DocItemLabel.DOCUMENT_INDEX,
             "otsl": DocItemLabel.TABLE,
             "section_header_level_1": DocItemLabel.SECTION_HEADER,
+            "section_header_level_2": DocItemLabel.SECTION_HEADER,
+            "section_header_level_3": DocItemLabel.SECTION_HEADER,
+            "section_header_level_4": DocItemLabel.SECTION_HEADER,
+            "section_header_level_5": DocItemLabel.SECTION_HEADER,
+            "section_header_level_6": DocItemLabel.SECTION_HEADER,
             "checkbox_selected": DocItemLabel.CHECKBOX_SELECTED,
             "checkbox_unselected": DocItemLabel.CHECKBOX_UNSELECTED,
             "text": DocItemLabel.TEXT,
@@ -3622,7 +3627,7 @@ class DoclingDocument(BaseModel):
                 rf"{DocItemLabel.PAGE_FOOTER}|{DocItemLabel.FORMULA}|"
                 rf"{DocItemLabel.CAPTION}|{DocItemLabel.PICTURE}|"
                 rf"{DocItemLabel.FOOTNOTE}|{DocItemLabel.CODE}|"
-                rf"{DocItemLabel.SECTION_HEADER}_level_1|"
+                rf"{DocItemLabel.SECTION_HEADER}_level_[1-6]|"
                 rf"{DocumentToken.ORDERED_LIST.value}|"
                 rf"{DocumentToken.UNORDERED_LIST.value}|"
                 rf"{DocItemLabel.KEY_VALUE_REGION}|"
@@ -3830,12 +3835,23 @@ class DoclingDocument(BaseModel):
                     if tag_name in [DocItemLabel.PAGE_HEADER, DocItemLabel.PAGE_FOOTER]:
                         content_layer = ContentLayer.FURNITURE
 
-                    doc.add_text(
-                        label=doc_label,
-                        text=text_content,
-                        prov=element_prov,
-                        content_layer=content_layer,
-                    )
+                    if doc_label == DocItemLabel.SECTION_HEADER:
+                        # Extract level from tag_name (e.g. "section_level_header_1" -> 1)
+                        level = int(tag_name.split("_")[-1])
+                        doc.add_heading(
+                            text=text_content,
+                            level=level,
+                            prov=element_prov,
+                            content_layer=content_layer,
+                        )
+                    else:
+                        doc.add_text(
+                            label=doc_label,
+                            text=text_content,
+                            prov=element_prov,
+                            content_layer=content_layer,
+                        )
+
         return doc
 
     @deprecated("Use save_as_doctags instead.")

{docling_core-2.33.0 → docling_core-2.33.1}/docling_core.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: docling-core
-Version: 2.33.0
+Version: 2.33.1
 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>
@@ -32,7 +32,7 @@ Requires-Dist: pandas<3.0.0,>=2.1.4
 Requires-Dist: pillow<12.0.0,>=10.0.0
 Requires-Dist: pyyaml<7.0.0,>=5.1
 Requires-Dist: typing-extensions<5.0.0,>=4.12.2
-Requires-Dist: typer<0.16.0,>=0.12.5
+Requires-Dist: typer<0.17.0,>=0.12.5
 Requires-Dist: latex2mathml<4.0.0,>=3.77.0
 Provides-Extra: chunking
 Requires-Dist: semchunk<3.0.0,>=2.2.0; extra == "chunking"

{docling_core-2.33.0 → docling_core-2.33.1}/docling_core.egg-info/requires.txt

@@ -6,7 +6,7 @@ pandas<3.0.0,>=2.1.4
 pillow<12.0.0,>=10.0.0
 pyyaml<7.0.0,>=5.1
 typing-extensions<5.0.0,>=4.12.2
-typer<0.16.0,>=0.12.5
+typer<0.17.0,>=0.12.5
 latex2mathml<4.0.0,>=3.77.0
 
 [chunking]

{docling_core-2.33.0 → docling_core-2.33.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "docling-core"
-version = "2.33.0"  # DO NOT EDIT, updated automatically
+version = "2.33.1"  # DO NOT EDIT, updated automatically
 description = "A python library to define and validate data types in Docling."
 license = "MIT"
 license-files = ["LICENSE"]
@@ -43,7 +43,7 @@ dependencies = [
     'pillow (>=10.0.0,<12.0.0)',
     'pyyaml (>=5.1,<7.0.0)',
     'typing-extensions (>=4.12.2,<5.0.0)',
-    'typer (>=0.12.5,<0.16.0)',
+    'typer (>=0.12.5,<0.17.0)',
     'latex2mathml (>=3.77.0,<4.0.0)',
 ]