docling-core 2.3.1__py3-none-any.whl → 2.4.0__py3-none-any.whl

docling_core/types/doc/document.py

@@ -551,6 +551,28 @@ class DocItem(
 
         return location
 
+    def get_image(self, doc: "DoclingDocument") -> Optional[PILImage.Image]:
+        """Returns the image of this DocItem.
+
+        The function returns None if this DocItem has no valid provenance or
+        if a valid image of the page containing this DocItem is not available
+        in doc.
+        """
+        if not len(self.prov):
+            return None
+
+        page = doc.pages.get(self.prov[0].page_no)
+        if page is None or page.size is None or page.image is None:
+            return None
+
+        page_image = page.image.pil_image
+        crop_bbox = (
+            self.prov[0]
+            .bbox.to_top_left_origin(page_height=page.size.height)
+            .scaled(scale=page_image.height / page.size.height)
+        )
+        return page_image.crop(crop_bbox.as_tuple())
+
 
 class TextItem(DocItem):
     """TextItem."""
@@ -633,6 +655,20 @@ class FloatingItem(DocItem):
             text += cap.resolve(doc).text
         return text
 
+    def get_image(self, doc: "DoclingDocument") -> Optional[PILImage.Image]:
+        """Returns the image corresponding to this FloatingItem.
+
+        This function returns the PIL image from self.image if one is available.
+        Otherwise, it uses DocItem.get_image to get an image of this FloatingItem.
+
+        In particular, when self.image is None, the function returns None if this
+        FloatingItem has no valid provenance or the doc does not contain a valid image
+        for the required page.
+        """
+        if self.image is not None:
+            return self.image.pil_image
+        return super().get_image(doc=doc)
+
 
 class PictureItem(FloatingItem):
     """PictureItem."""
@@ -1255,7 +1291,10 @@ class DoclingDocument(BaseModel):
                 # If the child is a NodeItem, recursively traverse it
                 if not isinstance(child, PictureItem) or traverse_pictures:
                     yield from self.iterate_items(
-                        child, _level=_level + 1, with_groups=with_groups
+                        child,
+                        _level=_level + 1,
+                        with_groups=with_groups,
+                        page_no=page_no,
                     )
 
     def print_element_tree(self):
@@ -1281,11 +1320,12 @@ class DoclingDocument(BaseModel):
         image_mode: ImageRefMode = ImageRefMode.PLACEHOLDER,
         indent: int = 4,
         text_width: int = -1,
+        page_no: Optional[int] = None,
     ) -> 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 +1334,9 @@ 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 +1344,6 @@ 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.
@@ -1320,7 +1357,7 @@ class DoclingDocument(BaseModel):
         in_list = False  # Track if we're currently processing list items
 
         for ix, (item, level) in enumerate(
-            self.iterate_items(self.body, with_groups=True)
+            self.iterate_items(self.body, with_groups=True, page_no=page_no)
         ):
             # If we've moved to a lower level, we're exiting one or more groups
             if level < previous_level:
@@ -1331,7 +1368,7 @@ 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-2.3.1.dist-info → docling_core-2.4.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling-core
-Version: 2.3.1
+Version: 2.4.0
 Summary: A python library to define and validate data types in Docling.
 Home-page: https://ds4sd.github.io/
 License: MIT

{docling_core-2.3.1.dist-info → docling_core-2.4.0.dist-info}/RECORD

@@ -21,7 +21,7 @@ docling_core/types/__init__.py,sha256=MVRSgsk5focwGyAplh_TRR3dEecIXpd98g_u3zZ5HX
 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=6KeHY4yl4Ry5nT6wacb8ujJ5LnyEZohXG5MAGhoPWGY,57771
 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-2.3.1.dist-info/LICENSE,sha256=2M9-6EoQ1sxFztTOkXGAtwUDJvnWaAHdB9BYWVwGkIw,1087
-docling_core-2.3.1.dist-info/METADATA,sha256=mASC44D6AB2bIACFr2oGrsZHtHRzn5e1wjBJyy6ccns,5432
-docling_core-2.3.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling_core-2.3.1.dist-info/entry_points.txt,sha256=jIxlWv3tnO04irlZc0zfhqJIgz1bg9Hha4AkaLWSdUA,177
-docling_core-2.3.1.dist-info/RECORD,,
+docling_core-2.4.0.dist-info/LICENSE,sha256=2M9-6EoQ1sxFztTOkXGAtwUDJvnWaAHdB9BYWVwGkIw,1087
+docling_core-2.4.0.dist-info/METADATA,sha256=fXFVK6Ey5DC15uSYgMixUmGxH6hEM-Kx06tK7jvW2IA,5432
+docling_core-2.4.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling_core-2.4.0.dist-info/entry_points.txt,sha256=jIxlWv3tnO04irlZc0zfhqJIgz1bg9Hha4AkaLWSdUA,177
+docling_core-2.4.0.dist-info/RECORD,,