docling 1.10.0__py3-none-any.whl → 1.11.0__py3-none-any.whl

docling/datamodel/document.py

@@ -11,6 +11,7 @@ from docling_core.types import FileInfoObject as DsFileInfoObject
 from docling_core.types import PageDimensions, PageReference, Prov, Ref
 from docling_core.types import Table as DsSchemaTable
 from docling_core.types import TableCell
+from docling_core.types.doc.base import Figure
 from pydantic import BaseModel
 from typing_extensions import deprecated
 
@@ -279,7 +280,7 @@ class ConvertedDocument(BaseModel):
                     ),
                 )
                 figures.append(
-                    BaseCell(
+                    Figure(
                         prov=[
                             Prov(
                                 bbox=target_bbox,
@@ -312,8 +313,76 @@ class ConvertedDocument(BaseModel):
     def render_as_dict(self):
         return self.output.model_dump(by_alias=True, exclude_none=True)
 
-    def render_as_markdown(self):
-        return self.output.export_to_markdown()
+    def render_as_markdown(
+        self,
+        delim: str = "\n\n",
+        main_text_start: int = 0,
+        main_text_stop: Optional[int] = None,
+        main_text_labels: list[str] = [
+            "title",
+            "subtitle-level-1",
+            "paragraph",
+            "caption",
+            "table",
+        ],
+        strict_text: bool = False,
+    ):
+        return self.output.export_to_markdown(
+            delim=delim,
+            main_text_start=main_text_start,
+            main_text_stop=main_text_stop,
+            main_text_labels=main_text_labels,
+            strict_text=strict_text,
+        )
+
+    def render_as_text(
+        self,
+        delim: str = "\n\n",
+        main_text_start: int = 0,
+        main_text_stop: Optional[int] = None,
+        main_text_labels: list[str] = [
+            "title",
+            "subtitle-level-1",
+            "paragraph",
+            "caption",
+        ],
+    ):
+        return self.output.export_to_markdown(
+            delim=delim,
+            main_text_start=main_text_start,
+            main_text_stop=main_text_stop,
+            main_text_labels=main_text_labels,
+            strict_text=True,
+        )
+
+    def render_as_doctags(
+        self,
+        delim: str = "\n\n",
+        main_text_start: int = 0,
+        main_text_stop: Optional[int] = None,
+        main_text_labels: list[str] = [
+            "title",
+            "subtitle-level-1",
+            "paragraph",
+            "caption",
+            "table",
+            "figure",
+        ],
+        page_tagging: bool = True,
+        location_tagging: bool = True,
+        location_dimensions: Tuple[int, int] = (100, 100),
+        add_new_line: bool = True,
+    ) -> str:
+        return self.output.export_to_document_tokens(
+            delim=delim,
+            main_text_start=main_text_start,
+            main_text_stop=main_text_stop,
+            main_text_labels=main_text_labels,
+            page_tagging=page_tagging,
+            location_tagging=location_tagging,
+            location_dimensions=location_dimensions,
+            add_new_line=add_new_line,
+        )
 
     def render_element_images(
         self, element_types: Tuple[PageElement] = (FigureElement,)

docling/utils/export.py

@@ -163,8 +163,12 @@ def generate_multimodal_pages(
         content_md = doc.export_to_markdown(
             main_text_start=start_ix, main_text_stop=end_ix
         )
+        # No page-tagging since we only do 1 page at the time
+        content_dt = doc.export_to_document_tokens(
+            main_text_start=start_ix, main_text_stop=end_ix, page_tagging=False
+        )
 
-        return content_text, content_md, page_cells, page_segments, page
+        return content_text, content_md, content_dt, page_cells, page_segments, page
 
     for ix, orig_item in enumerate(doc.main_text):
 

{docling-1.10.0.dist-info → docling-1.11.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 1.10.0
+Version: 1.11.0
 Summary: Docling PDF conversion package
 Home-page: https://github.com/DS4SD/docling
 License: MIT
@@ -21,8 +21,8 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Dist: certifi (>=2024.7.4)
 Requires-Dist: deepsearch-glm (>=0.21.0,<0.22.0)
-Requires-Dist: docling-core (>=1.1.3,<2.0.0)
-Requires-Dist: docling-ibm-models (>=1.1.3,<2.0.0)
+Requires-Dist: docling-core (>=1.2.0,<2.0.0)
+Requires-Dist: docling-ibm-models (>=1.1.7,<2.0.0)
 Requires-Dist: docling-parse (>=1.2.0,<2.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0)
 Requires-Dist: filetype (>=1.2.0,<2.0.0)

{docling-1.10.0.dist-info → docling-1.11.0.dist-info}/RECORD

@@ -5,7 +5,7 @@ docling/backend/docling_parse_backend.py,sha256=r3aJwsWR7qG47ElhOa9iQJJQauHMt950
 docling/backend/pypdfium2_backend.py,sha256=FggVFitmyMMmLar6vk6XQsavGOPQx95TD14opWYRMAY,8837
 docling/datamodel/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/datamodel/base_models.py,sha256=PSJe_Qlh2VJfijg3kkXOOqZbi_uqRHCmLjX__c5Buck,9155
-docling/datamodel/document.py,sha256=cG9RuAkFXCCGZqCHmhUtYeOA5PV6gjO3Y4i5lf2IM6I,13649
+docling/datamodel/document.py,sha256=oXPitPRd9Gyi7ZU4kfEc4K9eMVtTJDx1T-ellTwF3Ak,15716
 docling/datamodel/settings.py,sha256=t5g6wrEJnPa9gBzMMl8ppgBRUYz-8xgopEtfMS0ZH28,733
 docling/document_converter.py,sha256=5OiNafoaVcQhZ8ATF69xRp2KyFyKeSMhmwEFUoCzP-k,10980
 docling/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -19,10 +19,10 @@ docling/pipeline/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,
 docling/pipeline/base_model_pipeline.py,sha256=AC5NTR0xLy5JIZqsTINkKEHeCPqpyvJpuE_bcnZhyvI,529
 docling/pipeline/standard_model_pipeline.py,sha256=UTjyaEXvz9htYZz-IMTkn11cZwNjgvo_Fl2dfBVnRQs,1442
 docling/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-docling/utils/export.py,sha256=gP8609DtHp6bNGPhYpwe0g3J4qvc2HqQpHZnfl7hQZQ,5899
+docling/utils/export.py,sha256=ltPhhruS8sulHTYW0Rtjfc1I9lW3oH6QAF0oYewkz7k,6115
 docling/utils/layout_utils.py,sha256=FOFbL0hKzUoWXdZaeUvEtFqKv0IkPifIr4sdGW4suKs,31804
 docling/utils/utils.py,sha256=llhXSbIDNZ1MHOwBEfLHBAoJIAYI7QlPIonlI1jLUJ0,1208
-docling-1.10.0.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
-docling-1.10.0.dist-info/METADATA,sha256=1itpZzvKAruLgF_xPYhSFhqpUySogjDjT5u1HG2sGgM,8231
-docling-1.10.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling-1.10.0.dist-info/RECORD,,
+docling-1.11.0.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
+docling-1.11.0.dist-info/METADATA,sha256=lDqzdtE1ohikNmN3eyPQ31Qa30x9F5XN6FUTkTNGU9s,8231
+docling-1.11.0.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling-1.11.0.dist-info/RECORD,,