docling-core 2.3.0__py3-none-any.whl → 2.3.2__py3-none-any.whl

docling_core/transforms/chunker/hierarchical_chunker.py

@@ -183,14 +183,15 @@ class HierarchicalChunker(BaseChunker):
                         )
                         list_items = []  # reset
 
-                if isinstance(
-                    item, SectionHeaderItem
-                ) or (  # TODO remove when all captured as SectionHeaderItem:
+                if isinstance(item, SectionHeaderItem) or (
                     isinstance(item, TextItem)
-                    and item.label == DocItemLabel.SECTION_HEADER
+                    and item.label in [DocItemLabel.SECTION_HEADER, DocItemLabel.TITLE]
                 ):
-                    # TODO second branch not needed once cleanup above complete:
-                    level = item.level if isinstance(item, SectionHeaderItem) else 1
+                    level = (
+                        item.level
+                        if isinstance(item, SectionHeaderItem)
+                        else (0 if item.label == DocItemLabel.TITLE else 1)
+                    )
                     heading_by_level[level] = item.text
 
                     # remove headings of higher level as they just went out of scope

docling_core/types/doc/document.py

@@ -1284,8 +1284,8 @@ class DoclingDocument(BaseModel):
     ) -> str:
         r"""Serialize to Markdown.
 
-        Operates on a slice of the document's main_text as defined through arguments
-        main_text_start and main_text_stop; defaulting to the whole main_text.
+        Operates on a slice of the document's body as defined through arguments
+        from_element and to_element; defaulting to the whole document.
 
         :param delim: Delimiter to use when concatenating the various
                 Markdown parts. Defaults to "\n\n".
@@ -1294,11 +1294,9 @@ class DoclingDocument(BaseModel):
                 Defaults to 0.
         :type from_element: int
         :param to_element: Body slicing stop index
-                (exclusive). Defaults to None.
-        :type to_element: Optional[int]
+                (exclusive). Defaults to 0maxint.
+        :type to_element: int
         :param delim: str:  (Default value = "\n\n")
-        :param from_element: int:  (Default value = 0)
-        :param to_element: Optional[int]:  (Default value = None)
         :param labels: set[DocItemLabel]
         :param "subtitle-level-1":
         :param "paragraph":
@@ -1306,7 +1304,6 @@ class DoclingDocument(BaseModel):
         :param "table":
         :param "Text":
         :param "text":
-        :param ]:
         :param strict_text: bool:  (Default value = False)
         :param image_placeholder str:  (Default value = "<!-- image -->")
             the placeholder to include to position images in the markdown.
@@ -1331,7 +1328,7 @@ class DoclingDocument(BaseModel):
 
             previous_level = level  # Update previous_level for next iteration
 
-            if ix < from_element and to_element <= ix:
+            if ix < from_element or to_element <= ix:
                 continue  # skip as many items as you want
 
             # Handle newlines between different types of content

{docling_core-2.3.0.dist-info → docling_core-2.3.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling-core
-Version: 2.3.0
+Version: 2.3.2
 Summary: A python library to define and validate data types in Docling.
 Home-page: https://ds4sd.github.io/
 License: MIT

{docling_core-2.3.0.dist-info → docling_core-2.3.2.dist-info}/RECORD

@@ -16,12 +16,12 @@ 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=uG7nNoUCFqWeQAKydQg731JYJ9sayUe4J48nMF0VHE8,8097
+docling_core/transforms/chunker/hierarchical_chunker.py,sha256=V4FiOYqL0GgBqVB7x6CafAJs3WF5oYifKIiexVggGPE,8086
 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=XF43-v9oflV-E5r2k2quoKvq8qBp5mAB_VunshY9b10,56356
+docling_core/types/doc/document.py,sha256=ED-x3Hc0E3gEYZBt3ue3auYhiuEegRfz9_cq-LJEThE,56196
 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
@@ -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-2.3.0.dist-info/LICENSE,sha256=2M9-6EoQ1sxFztTOkXGAtwUDJvnWaAHdB9BYWVwGkIw,1087
-docling_core-2.3.0.dist-info/METADATA,sha256=1cthKA2Ke9tujdMQzh4sNnBB1K90yFNQ9TCvD2MhKDI,5432
-docling_core-2.3.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling_core-2.3.0.dist-info/entry_points.txt,sha256=jIxlWv3tnO04irlZc0zfhqJIgz1bg9Hha4AkaLWSdUA,177
-docling_core-2.3.0.dist-info/RECORD,,
+docling_core-2.3.2.dist-info/LICENSE,sha256=2M9-6EoQ1sxFztTOkXGAtwUDJvnWaAHdB9BYWVwGkIw,1087
+docling_core-2.3.2.dist-info/METADATA,sha256=VUY20vSISoTRU7vdkP-mAlInUVWyqyGURfIfmqTeGdA,5432
+docling_core-2.3.2.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling_core-2.3.2.dist-info/entry_points.txt,sha256=jIxlWv3tnO04irlZc0zfhqJIgz1bg9Hha4AkaLWSdUA,177
+docling_core-2.3.2.dist-info/RECORD,,