docling 0.1.2__tar.gz → 1.5.0__tar.gz

docling-1.5.0/PKG-INFO

@@ -0,0 +1,192 @@
+Metadata-Version: 2.1
+Name: docling
+Version: 1.5.0
+Summary: Docling PDF conversion package
+Home-page: https://github.com/DS4SD/docling
+License: MIT
+Keywords: docling,convert,document,pdf,layout model,segmentation,table structure,table former
+Author: Christoph Auer
+Author-email: cau@zurich.ibm.com
+Requires-Python: >=3.10,<4.0
+Classifier: Development Status :: 5 - Production/Stable
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Operating System :: MacOS :: MacOS X
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Provides-Extra: easyocr
+Provides-Extra: ocr
+Requires-Dist: certifi (>=2024.7.4)
+Requires-Dist: deepsearch-glm (>=0.19.0,<1)
+Requires-Dist: docling-core (>=1.1.2,<2.0.0)
+Requires-Dist: docling-ibm-models (>=1.1.1,<2.0.0)
+Requires-Dist: docling-parse (>=0.2.0,<0.3.0)
+Requires-Dist: easyocr (>=1.7,<2.0) ; extra == "easyocr" or extra == "ocr"
+Requires-Dist: filetype (>=1.2.0,<2.0.0)
+Requires-Dist: huggingface_hub (>=0.23,<1)
+Requires-Dist: pydantic (>=2.0.0,<3.0.0)
+Requires-Dist: pydantic-settings (>=2.3.0,<3.0.0)
+Requires-Dist: pypdfium2 (>=4.30.0,<5.0.0)
+Requires-Dist: requests (>=2.32.3,<3.0.0)
+Project-URL: Repository, https://github.com/DS4SD/docling
+Description-Content-Type: text/markdown
+
+<p align="center">
+  <a href="https://github.com/ds4sd/docling">
+    <img loading="lazy" alt="Docling" src="https://github.com/DS4SD/docling/raw/main/logo.png" width="150" />
+  </a>
+</p>
+
+# Docling
+
+[![arXiv](https://img.shields.io/badge/arXiv-2408.09869-b31b1b.svg)](https://arxiv.org/abs/2408.09869)
+[![PyPI version](https://img.shields.io/pypi/v/docling)](https://pypi.org/project/docling/)
+![Python](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-blue)
+[![Poetry](https://img.shields.io/endpoint?url=https://python-poetry.org/badge/v0.json)](https://python-poetry.org/)
+[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
+[![Imports: isort](https://img.shields.io/badge/%20imports-isort-%231674b1?style=flat&labelColor=ef8336)](https://pycqa.github.io/isort/)
+[![Pydantic v2](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/pydantic/pydantic/main/docs/badge/v2.json)](https://pydantic.dev)
+[![pre-commit](https://img.shields.io/badge/pre--commit-enabled-brightgreen?logo=pre-commit&logoColor=white)](https://github.com/pre-commit/pre-commit)
+[![License MIT](https://img.shields.io/github/license/DS4SD/docling)](https://opensource.org/licenses/MIT)
+
+Docling bundles PDF document conversion to JSON and Markdown in an easy, self-contained package.
+
+## Features
+* ⚡ Converts any PDF document to JSON or Markdown format, stable and lightning fast
+* 📑 Understands detailed page layout, reading order and recovers table structures
+* 📝 Extracts metadata from the document, such as title, authors, references and language
+* 🔍 Optionally applies OCR (use with scanned PDFs)
+
+## Installation
+
+To use Docling, simply install `docling` from your package manager, e.g. pip:
+```bash
+pip install docling
+```
+
+> [!NOTE]
+> Works on macOS and Linux environments. Windows platforms are currently not tested.
+
+### Development setup
+
+To develop for Docling, you need Python 3.10 / 3.11 / 3.12 and Poetry. You can then install from your local clone's root dir:
+```bash
+poetry install --all-extras
+```
+
+## Usage
+
+### Convert a single document
+
+To convert invidual PDF documents, use `convert_single()`, for example:
+```python
+from docling.document_converter import DocumentConverter
+
+source = "https://arxiv.org/pdf/2206.01062"  # PDF path or URL
+converter = DocumentConverter()
+doc = converter.convert_single(source)
+print(doc.export_to_markdown())  # output: "## DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis [...]"
+```
+
+### Convert a batch of documents
+
+For an example of batch-converting documents, see [batch_convert.py](https://github.com/DS4SD/docling/blob/main/examples/batch_convert.py).
+
+From a local repo clone, you can run it with:
+
+```
+python examples/batch_convert.py
+```
+The output of the above command will be written to `./scratch`.
+
+### Adjust pipeline features
+
+The example file [custom_convert.py](https://github.com/DS4SD/docling/blob/main/examples/custom_convert.py) contains multiple ways
+one can adjust the conversion pipeline and features.
+
+
+#### Control pipeline options
+
+You can control if table structure recognition or OCR should be performed by arguments passed to `DocumentConverter`:
+```python
+doc_converter = DocumentConverter(
+    artifacts_path=artifacts_path,
+    pipeline_options=PipelineOptions(
+        do_table_structure=False,  # controls if table structure is recovered
+        do_ocr=True,  # controls if OCR is applied (ignores programmatic content)
+    ),
+)
+```
+
+#### Control table extraction options
+
+You can control if table structure recognition should map the recognized structure back to PDF cells (default) or use text cells from the structure prediction itself.
+This can improve output quality if you find that multiple columns in extracted tables are erroneously merged into one.
+
+
+```python
+pipeline_options = PipelineOptions(do_table_structure=True)
+pipeline_options.table_structure_options.do_cell_matching = False  # uses text cells predicted from table structure model
+
+doc_converter = DocumentConverter(
+    artifacts_path=artifacts_path,
+    pipeline_options=pipeline_options,
+)
+```
+
+### Impose limits on the document size
+
+You can limit the file size and number of pages which should be allowed to process per document:
+```python
+conv_input = DocumentConversionInput.from_paths(
+    paths=[Path("./test/data/2206.01062.pdf")],
+    limits=DocumentLimits(max_num_pages=100, max_file_size=20971520)
+)
+```
+
+### Convert from binary PDF streams
+
+You can convert PDFs from a binary stream instead of from the filesystem as follows:
+```python
+buf = BytesIO(your_binary_stream)
+docs = [DocumentStream(filename="my_doc.pdf", stream=buf)]
+conv_input = DocumentConversionInput.from_streams(docs)
+converted_docs = doc_converter.convert(conv_input)
+```
+### Limit resource usage
+
+You can limit the CPU threads used by Docling by setting the environment variable `OMP_NUM_THREADS` accordingly. The default setting is using 4 CPU threads.
+
+
+## Contributing
+
+Please read [Contributing to Docling](https://github.com/DS4SD/docling/blob/main/CONTRIBUTING.md) for details.
+
+
+## References
+
+If you use Docling in your projects, please consider citing the following:
+
+```bib
+@techreport{Docling,
+  author = {Deep Search Team},
+  month = {8},
+  title = {{Docling Technical Report}},
+  url={https://arxiv.org/abs/2408.09869},
+  eprint={2408.09869},
+  doi = "10.48550/arXiv.2408.09869",
+  version = {1.0.0},
+  year = {2024}
+}
+```
+
+## License
+
+The Docling codebase is under MIT license.
+For individual model usage, please refer to the model licenses found in the original packages.
+

docling-1.5.0/README.md

@@ -0,0 +1,153 @@
+<p align="center">
+  <a href="https://github.com/ds4sd/docling">
+    <img loading="lazy" alt="Docling" src="https://github.com/DS4SD/docling/raw/main/logo.png" width="150" />
+  </a>
+</p>
+
+# Docling
+
+[![arXiv](https://img.shields.io/badge/arXiv-2408.09869-b31b1b.svg)](https://arxiv.org/abs/2408.09869)
+[![PyPI version](https://img.shields.io/pypi/v/docling)](https://pypi.org/project/docling/)
+![Python](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-blue)
+[![Poetry](https://img.shields.io/endpoint?url=https://python-poetry.org/badge/v0.json)](https://python-poetry.org/)
+[![Code style: black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
+[![Imports: isort](https://img.shields.io/badge/%20imports-isort-%231674b1?style=flat&labelColor=ef8336)](https://pycqa.github.io/isort/)
+[![Pydantic v2](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/pydantic/pydantic/main/docs/badge/v2.json)](https://pydantic.dev)
+[![pre-commit](https://img.shields.io/badge/pre--commit-enabled-brightgreen?logo=pre-commit&logoColor=white)](https://github.com/pre-commit/pre-commit)
+[![License MIT](https://img.shields.io/github/license/DS4SD/docling)](https://opensource.org/licenses/MIT)
+
+Docling bundles PDF document conversion to JSON and Markdown in an easy, self-contained package.
+
+## Features
+* ⚡ Converts any PDF document to JSON or Markdown format, stable and lightning fast
+* 📑 Understands detailed page layout, reading order and recovers table structures
+* 📝 Extracts metadata from the document, such as title, authors, references and language
+* 🔍 Optionally applies OCR (use with scanned PDFs)
+
+## Installation
+
+To use Docling, simply install `docling` from your package manager, e.g. pip:
+```bash
+pip install docling
+```
+
+> [!NOTE]
+> Works on macOS and Linux environments. Windows platforms are currently not tested.
+
+### Development setup
+
+To develop for Docling, you need Python 3.10 / 3.11 / 3.12 and Poetry. You can then install from your local clone's root dir:
+```bash
+poetry install --all-extras
+```
+
+## Usage
+
+### Convert a single document
+
+To convert invidual PDF documents, use `convert_single()`, for example:
+```python
+from docling.document_converter import DocumentConverter
+
+source = "https://arxiv.org/pdf/2206.01062"  # PDF path or URL
+converter = DocumentConverter()
+doc = converter.convert_single(source)
+print(doc.export_to_markdown())  # output: "## DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis [...]"
+```
+
+### Convert a batch of documents
+
+For an example of batch-converting documents, see [batch_convert.py](https://github.com/DS4SD/docling/blob/main/examples/batch_convert.py).
+
+From a local repo clone, you can run it with:
+
+```
+python examples/batch_convert.py
+```
+The output of the above command will be written to `./scratch`.
+
+### Adjust pipeline features
+
+The example file [custom_convert.py](https://github.com/DS4SD/docling/blob/main/examples/custom_convert.py) contains multiple ways
+one can adjust the conversion pipeline and features.
+
+
+#### Control pipeline options
+
+You can control if table structure recognition or OCR should be performed by arguments passed to `DocumentConverter`:
+```python
+doc_converter = DocumentConverter(
+    artifacts_path=artifacts_path,
+    pipeline_options=PipelineOptions(
+        do_table_structure=False,  # controls if table structure is recovered
+        do_ocr=True,  # controls if OCR is applied (ignores programmatic content)
+    ),
+)
+```
+
+#### Control table extraction options
+
+You can control if table structure recognition should map the recognized structure back to PDF cells (default) or use text cells from the structure prediction itself.
+This can improve output quality if you find that multiple columns in extracted tables are erroneously merged into one.
+
+
+```python
+pipeline_options = PipelineOptions(do_table_structure=True)
+pipeline_options.table_structure_options.do_cell_matching = False  # uses text cells predicted from table structure model
+
+doc_converter = DocumentConverter(
+    artifacts_path=artifacts_path,
+    pipeline_options=pipeline_options,
+)
+```
+
+### Impose limits on the document size
+
+You can limit the file size and number of pages which should be allowed to process per document:
+```python
+conv_input = DocumentConversionInput.from_paths(
+    paths=[Path("./test/data/2206.01062.pdf")],
+    limits=DocumentLimits(max_num_pages=100, max_file_size=20971520)
+)
+```
+
+### Convert from binary PDF streams
+
+You can convert PDFs from a binary stream instead of from the filesystem as follows:
+```python
+buf = BytesIO(your_binary_stream)
+docs = [DocumentStream(filename="my_doc.pdf", stream=buf)]
+conv_input = DocumentConversionInput.from_streams(docs)
+converted_docs = doc_converter.convert(conv_input)
+```
+### Limit resource usage
+
+You can limit the CPU threads used by Docling by setting the environment variable `OMP_NUM_THREADS` accordingly. The default setting is using 4 CPU threads.
+
+
+## Contributing
+
+Please read [Contributing to Docling](https://github.com/DS4SD/docling/blob/main/CONTRIBUTING.md) for details.
+
+
+## References
+
+If you use Docling in your projects, please consider citing the following:
+
+```bib
+@techreport{Docling,
+  author = {Deep Search Team},
+  month = {8},
+  title = {{Docling Technical Report}},
+  url={https://arxiv.org/abs/2408.09869},
+  eprint={2408.09869},
+  doi = "10.48550/arXiv.2408.09869",
+  version = {1.0.0},
+  year = {2024}
+}
+```
+
+## License
+
+The Docling codebase is under MIT license.
+For individual model usage, please refer to the model licenses found in the original packages.

{docling-0.1.2 → docling-1.5.0}/docling/backend/abstract_backend.py

@@ -35,7 +35,7 @@ class PdfPageBackend(ABC):
 
 class PdfDocumentBackend(ABC):
     @abstractmethod
-    def __init__(self, path_or_stream: Iterable[Union[BytesIO, Path]]):
+    def __init__(self, path_or_stream: Union[BytesIO, Path]):
         pass
 
     @abstractmethod

docling-1.5.0/docling/backend/docling_parse_backend.py

@@ -0,0 +1,187 @@
+import logging
+import random
+import time
+from io import BytesIO
+from pathlib import Path
+from typing import Iterable, List, Optional, Union
+
+import pypdfium2 as pdfium
+from docling_parse.docling_parse import pdf_parser
+from PIL import Image, ImageDraw
+from pypdfium2 import PdfPage
+
+from docling.backend.abstract_backend import PdfDocumentBackend, PdfPageBackend
+from docling.datamodel.base_models import BoundingBox, Cell, CoordOrigin, PageSize
+
+_log = logging.getLogger(__name__)
+
+
+class DoclingParsePageBackend(PdfPageBackend):
+    def __init__(self, page_obj: PdfPage, docling_page_obj):
+        super().__init__(page_obj)
+        self._ppage = page_obj
+        self._dpage = docling_page_obj
+        self.text_page = None
+
+    def get_text_in_rect(self, bbox: BoundingBox) -> str:
+        # Find intersecting cells on the page
+        text_piece = ""
+        page_size = self.get_size()
+        parser_width = self._dpage["width"]
+        parser_height = self._dpage["height"]
+
+        scale = (
+            1  # FIX - Replace with param in get_text_in_rect across backends (optional)
+        )
+
+        for i in range(len(self._dpage["cells"])):
+            rect = self._dpage["cells"][i]["box"]["device"]
+            x0, y0, x1, y1 = rect
+            cell_bbox = BoundingBox(
+                l=x0 * scale * page_size.width / parser_width,
+                b=y0 * scale * page_size.height / parser_height,
+                r=x1 * scale * page_size.width / parser_width,
+                t=y1 * scale * page_size.height / parser_height,
+                coord_origin=CoordOrigin.BOTTOMLEFT,
+            ).to_top_left_origin(page_size.height * scale)
+
+            overlap_frac = cell_bbox.intersection_area_with(bbox) / cell_bbox.area()
+
+            if overlap_frac > 0.5:
+                if len(text_piece) > 0:
+                    text_piece += " "
+                text_piece += self._dpage["cells"][i]["content"]["rnormalized"]
+
+        return text_piece
+
+    def get_text_cells(self) -> Iterable[Cell]:
+        cells = []
+        cell_counter = 0
+
+        page_size = self.get_size()
+
+        parser_width = self._dpage["width"]
+        parser_height = self._dpage["height"]
+
+        for i in range(len(self._dpage["cells"])):
+            rect = self._dpage["cells"][i]["box"]["device"]
+            x0, y0, x1, y1 = rect
+            text_piece = self._dpage["cells"][i]["content"]["rnormalized"]
+            cells.append(
+                Cell(
+                    id=cell_counter,
+                    text=text_piece,
+                    bbox=BoundingBox(
+                        # l=x0, b=y0, r=x1, t=y1,
+                        l=x0 * page_size.width / parser_width,
+                        b=y0 * page_size.height / parser_height,
+                        r=x1 * page_size.width / parser_width,
+                        t=y1 * page_size.height / parser_height,
+                        coord_origin=CoordOrigin.BOTTOMLEFT,
+                    ).to_top_left_origin(page_size.height),
+                )
+            )
+            cell_counter += 1
+
+        def draw_clusters_and_cells():
+            image = (
+                self.get_page_image()
+            )  # make new image to avoid drawing on the saved ones
+            draw = ImageDraw.Draw(image)
+            for c in cells:
+                x0, y0, x1, y1 = c.bbox.as_tuple()
+                cell_color = (
+                    random.randint(30, 140),
+                    random.randint(30, 140),
+                    random.randint(30, 140),
+                )
+                draw.rectangle([(x0, y0), (x1, y1)], outline=cell_color)
+            image.show()
+
+        # before merge:
+        # draw_clusters_and_cells()
+
+        # cells = merge_horizontal_cells(cells)
+
+        # after merge:
+        # draw_clusters_and_cells()
+
+        return cells
+
+    def get_page_image(
+        self, scale: int = 1, cropbox: Optional[BoundingBox] = None
+    ) -> Image.Image:
+
+        page_size = self.get_size()
+
+        if not cropbox:
+            cropbox = BoundingBox(
+                l=0,
+                r=page_size.width,
+                t=0,
+                b=page_size.height,
+                coord_origin=CoordOrigin.TOPLEFT,
+            )
+            padbox = BoundingBox(
+                l=0, r=0, t=0, b=0, coord_origin=CoordOrigin.BOTTOMLEFT
+            )
+        else:
+            padbox = cropbox.to_bottom_left_origin(page_size.height)
+            padbox.r = page_size.width - padbox.r
+            padbox.t = page_size.height - padbox.t
+
+        image = (
+            self._ppage.render(
+                scale=scale * 1.5,
+                rotation=0,  # no additional rotation
+                crop=padbox.as_tuple(),
+            )
+            .to_pil()
+            .resize(size=(round(cropbox.width * scale), round(cropbox.height * scale)))
+        )  # We resize the image from 1.5x the given scale to make it sharper.
+
+        return image
+
+    def get_size(self) -> PageSize:
+        return PageSize(width=self._ppage.get_width(), height=self._ppage.get_height())
+
+    def unload(self):
+        self._ppage = None
+        self._dpage = None
+        self.text_page = None
+
+
+class DoclingParseDocumentBackend(PdfDocumentBackend):
+    def __init__(self, path_or_stream: Union[BytesIO, Path]):
+        super().__init__(path_or_stream)
+        self._pdoc = pdfium.PdfDocument(path_or_stream)
+        # Parsing cells with docling_parser call
+        parser = pdf_parser()
+
+        start_pb_time = time.time()
+
+        if isinstance(path_or_stream, BytesIO):
+            self._parser_doc = parser.find_cells_from_bytesio(path_or_stream)
+        else:
+            self._parser_doc = parser.find_cells(str(path_or_stream))
+
+        end_pb_time = time.time() - start_pb_time
+        _log.info(
+            f"Time to parse {path_or_stream.name} with docling-parse: time={end_pb_time:.3f}"
+        )
+
+    def page_count(self) -> int:
+        return len(self._parser_doc["pages"])
+
+    def load_page(self, page_no: int) -> PdfPage:
+        return DoclingParsePageBackend(
+            self._pdoc[page_no], self._parser_doc["pages"][page_no]
+        )
+
+    def is_valid(self) -> bool:
+        return self.page_count() > 0
+
+    def unload(self):
+        self._pdoc.close()
+        self._pdoc = None
+        self._parser_doc = None

{docling-0.1.2 → docling-1.5.0}/docling/backend/pypdfium2_backend.py

@@ -134,7 +134,9 @@ class PyPdfiumPageBackend(PdfPageBackend):
             return merged_cells
 
         def draw_clusters_and_cells():
-            image = self.get_page_image()
+            image = (
+                self.get_page_image()
+            )  # make new image to avoid drawing on the saved ones
             draw = ImageDraw.Draw(image)
             for c in cells:
                 x0, y0, x1, y1 = c.bbox.as_tuple()
@@ -199,15 +201,9 @@ class PyPdfiumPageBackend(PdfPageBackend):
 
 
 class PyPdfiumDocumentBackend(PdfDocumentBackend):
-    def __init__(self, path_or_stream: Iterable[Union[BytesIO, Path]]):
+    def __init__(self, path_or_stream: Union[BytesIO, Path]):
         super().__init__(path_or_stream)
-
-        if isinstance(path_or_stream, Path):
-            self._pdoc = pdfium.PdfDocument(path_or_stream)
-        elif isinstance(path_or_stream, BytesIO):
-            self._pdoc = pdfium.PdfDocument(
-                path_or_stream
-            )  # TODO Fix me, won't accept bytes.
+        self._pdoc = pdfium.PdfDocument(path_or_stream)
 
     def page_count(self) -> int:
         return len(self._pdoc)

{docling-0.1.2 → docling-1.5.0}/docling/datamodel/base_models.py

@@ -1,9 +1,12 @@
+import copy
+import warnings
 from enum import Enum, auto
 from io import BytesIO
-from typing import Any, Dict, List, Optional, Tuple, Union
+from typing import Annotated, Any, Dict, List, Optional, Tuple, Union
 
 from PIL.Image import Image
-from pydantic import BaseModel, ConfigDict, model_validator
+from pydantic import BaseModel, ConfigDict, Field, model_validator
+from typing_extensions import Self
 
 from docling.backend.abstract_backend import PdfPageBackend
 
@@ -47,6 +50,15 @@ class BoundingBox(BaseModel):
     def height(self):
         return abs(self.t - self.b)
 
+    def scaled(self, scale: float) -> "BoundingBox":
+        out_bbox = copy.deepcopy(self)
+        out_bbox.l *= scale
+        out_bbox.r *= scale
+        out_bbox.t *= scale
+        out_bbox.b *= scale
+
+        return out_bbox
+
     def as_tuple(self):
         if self.coord_origin == CoordOrigin.TOPLEFT:
             return (self.l, self.t, self.r, self.b)
@@ -180,8 +192,7 @@ class TableStructurePrediction(BaseModel):
     table_map: Dict[int, TableElement] = {}
 
 
-class TextElement(BasePageElement):
-    ...
+class TextElement(BasePageElement): ...
 
 
 class FigureData(BaseModel):
@@ -225,14 +236,30 @@ class Page(BaseModel):
     model_config = ConfigDict(arbitrary_types_allowed=True)
 
     page_no: int
-    page_hash: str = None
-    size: PageSize = None
-    image: Image = None
+    page_hash: Optional[str] = None
+    size: Optional[PageSize] = None
     cells: List[Cell] = None
     predictions: PagePredictions = PagePredictions()
-    assembled: AssembledUnit = None
+    assembled: Optional[AssembledUnit] = None
+
+    _backend: Optional[PdfPageBackend] = (
+        None  # Internal PDF backend. By default it is cleared during assembling.
+    )
+    _default_image_scale: float = 1.0  # Default image scale for external usage.
+    _image_cache: Dict[float, Image] = (
+        {}
+    )  # Cache of images in different scales. By default it is cleared during assembling.
+
+    def get_image(self, scale: float = 1.0) -> Optional[Image]:
+        if self._backend is None:
+            return self._image_cache.get(scale, None)
+        if not scale in self._image_cache:
+            self._image_cache[scale] = self._backend.get_page_image(scale=scale)
+        return self._image_cache[scale]
 
-    _backend: PdfPageBackend = None  # Internal PDF backend
+    @property
+    def image(self) -> Optional[Image]:
+        return self.get_image(scale=self._default_image_scale)
 
 
 class DocumentStream(BaseModel):
@@ -242,6 +269,36 @@ class DocumentStream(BaseModel):
     stream: BytesIO
 
 
+class TableStructureOptions(BaseModel):
+    do_cell_matching: bool = (
+        True
+        # True:  Matches predictions back to PDF cells. Can break table output if PDF cells
+        #        are merged across table columns.
+        # False: Let table structure model define the text cells, ignore PDF cells.
+    )
+
+
 class PipelineOptions(BaseModel):
-    do_table_structure: bool = True
-    do_ocr: bool = False
+    do_table_structure: bool = True  # True: perform table structure extraction
+    do_ocr: bool = False  # True: perform OCR, replace programmatic PDF text
+
+    table_structure_options: TableStructureOptions = TableStructureOptions()
+
+
+class AssembleOptions(BaseModel):
+    keep_page_images: Annotated[
+        bool,
+        Field(
+            deprecated="`keep_page_images` is depreacted, set the value of `images_scale` instead"
+        ),
+    ] = False  # False: page images are removed in the assemble step
+    images_scale: Optional[float] = None  # if set, the scale for generated images
+
+    @model_validator(mode="after")
+    def set_page_images_from_deprecated(self) -> Self:
+        with warnings.catch_warnings():
+            warnings.simplefilter("ignore", DeprecationWarning)
+            default_scale = 1.0
+            if self.keep_page_images and self.images_scale is None:
+                self.images_scale = default_scale
+        return self