docling 2.7.1__py3-none-any.whl → 2.8.1__py3-none-any.whl

docling/backend/msword_backend.py

@@ -507,18 +507,19 @@ class MsWordDocumentBackend(DeclarativeDocumentBackend):
 
         image_data = get_docx_image(element, drawing_blip)
         image_bytes = BytesIO(image_data)
+        level = self.get_level()
         # Open the BytesIO object with PIL to create an Image
         try:
             pil_image = Image.open(image_bytes)
             doc.add_picture(
-                parent=self.parents[self.level],
+                parent=self.parents[level - 1],
                 image=ImageRef.from_pil(image=pil_image, dpi=72),
                 caption=None,
             )
         except (UnidentifiedImageError, OSError) as e:
             _log.warning("Warning: image cannot be loaded by Pillow")
             doc.add_picture(
-                parent=self.parents[self.level],
+                parent=self.parents[level - 1],
                 caption=None,
             )
         return

docling/cli/main.py

@@ -27,10 +27,12 @@ from docling.datamodel.pipeline_options import (
     OcrMacOptions,
     OcrOptions,
     PdfPipelineOptions,
+    RapidOcrOptions,
     TableFormerMode,
     TesseractCliOcrOptions,
     TesseractOcrOptions,
 )
+from docling.datamodel.settings import settings
 from docling.document_converter import DocumentConverter, FormatOption, PdfFormatOption
 
 warnings.filterwarnings(action="ignore", category=UserWarning, module="pydantic|torch")
@@ -76,6 +78,7 @@ class OcrEngine(str, Enum):
     TESSERACT_CLI = "tesseract_cli"
     TESSERACT = "tesseract"
     OCRMAC = "ocrmac"
+    RAPIDOCR = "rapidocr"
 
 
 def export_documents(
@@ -210,6 +213,24 @@ def convert(
             help="Set the verbosity level. -v for info logging, -vv for debug logging.",
         ),
     ] = 0,
+    debug_visualize_cells: Annotated[
+        bool,
+        typer.Option(..., help="Enable debug output which visualizes the PDF cells"),
+    ] = False,
+    debug_visualize_ocr: Annotated[
+        bool,
+        typer.Option(..., help="Enable debug output which visualizes the OCR cells"),
+    ] = False,
+    debug_visualize_layout: Annotated[
+        bool,
+        typer.Option(
+            ..., help="Enable debug output which visualizes the layour clusters"
+        ),
+    ] = False,
+    debug_visualize_tables: Annotated[
+        bool,
+        typer.Option(..., help="Enable debug output which visualizes the table cells"),
+    ] = False,
     version: Annotated[
         Optional[bool],
         typer.Option(
@@ -227,6 +248,11 @@ def convert(
     elif verbose == 2:
         logging.basicConfig(level=logging.DEBUG)
 
+    settings.debug.visualize_cells = debug_visualize_cells
+    settings.debug.visualize_layout = debug_visualize_layout
+    settings.debug.visualize_tables = debug_visualize_tables
+    settings.debug.visualize_ocr = debug_visualize_ocr
+
     if from_formats is None:
         from_formats = [e for e in InputFormat]
 
@@ -262,6 +288,8 @@ def convert(
         ocr_options = TesseractOcrOptions(force_full_page_ocr=force_ocr)
     elif ocr_engine == OcrEngine.OCRMAC:
         ocr_options = OcrMacOptions(force_full_page_ocr=force_ocr)
+    elif ocr_engine == OcrEngine.RAPIDOCR:
+        ocr_options = RapidOcrOptions(force_full_page_ocr=force_ocr)
     else:
         raise RuntimeError(f"Unexpected OCR engine type {ocr_engine}")
 

docling/datamodel/pipeline_options.py

@@ -29,6 +29,42 @@ class OcrOptions(BaseModel):
     )
 
 
+class RapidOcrOptions(OcrOptions):
+    kind: Literal["rapidocr"] = "rapidocr"
+
+    # English and chinese are the most commly used models and have been tested with RapidOCR.
+    lang: List[str] = [
+        "english",
+        "chinese",
+    ]  # However, language as a parameter is not supported by rapidocr yet and hence changing this options doesn't affect anything.
+    # For more details on supported languages by RapidOCR visit https://rapidai.github.io/RapidOCRDocs/blog/2022/09/28/%E6%94%AF%E6%8C%81%E8%AF%86%E5%88%AB%E8%AF%AD%E8%A8%80/
+
+    # For more details on the following options visit https://rapidai.github.io/RapidOCRDocs/install_usage/api/RapidOCR/
+    text_score: float = 0.5  # same default as rapidocr
+
+    use_det: Optional[bool] = None  # same default as rapidocr
+    use_cls: Optional[bool] = None  # same default as rapidocr
+    use_rec: Optional[bool] = None  # same default as rapidocr
+
+    # class Device(Enum):
+    #     CPU = "CPU"
+    #     CUDA = "CUDA"
+    #     DIRECTML = "DIRECTML"
+    #     AUTO = "AUTO"
+
+    # device: Device = Device.AUTO  # Default value is AUTO
+
+    print_verbose: bool = False  # same default as rapidocr
+
+    det_model_path: Optional[str] = None  # same default as rapidocr
+    cls_model_path: Optional[str] = None  # same default as rapidocr
+    rec_model_path: Optional[str] = None  # same default as rapidocr
+
+    model_config = ConfigDict(
+        extra="forbid",
+    )
+
+
 class EasyOcrOptions(OcrOptions):
     kind: Literal["easyocr"] = "easyocr"
     lang: List[str] = ["fr", "de", "es", "en"]

docling/models/rapid_ocr_model.py

@@ -0,0 +1,147 @@
+import logging
+from typing import Iterable
+
+import numpy
+from docling_core.types.doc import BoundingBox, CoordOrigin
+
+from docling.datamodel.base_models import OcrCell, Page
+from docling.datamodel.document import ConversionResult
+from docling.datamodel.pipeline_options import RapidOcrOptions
+from docling.datamodel.settings import settings
+from docling.models.base_ocr_model import BaseOcrModel
+from docling.utils.profiling import TimeRecorder
+
+_log = logging.getLogger(__name__)
+
+
+class RapidOcrModel(BaseOcrModel):
+    def __init__(self, enabled: bool, options: RapidOcrOptions):
+        super().__init__(enabled=enabled, options=options)
+        self.options: RapidOcrOptions
+
+        self.scale = 3  # multiplier for 72 dpi == 216 dpi.
+
+        if self.enabled:
+            try:
+                from rapidocr_onnxruntime import RapidOCR  # type: ignore
+            except ImportError:
+                raise ImportError(
+                    "RapidOCR is not installed. Please install it via `pip install rapidocr_onnxruntime` to use this OCR engine. "
+                    "Alternatively, Docling has support for other OCR engines. See the documentation."
+                )
+
+            # This configuration option will be revamped while introducing device settings for all models.
+            # For the moment we will default to auto and let onnx-runtime pick the best.
+            cls_use_cuda = True
+            rec_use_cuda = True
+            det_use_cuda = True
+            det_use_dml = True
+            cls_use_dml = True
+            rec_use_dml = True
+
+            # # Same as Defaults in RapidOCR
+            # cls_use_cuda = False
+            # rec_use_cuda = False
+            # det_use_cuda = False
+            # det_use_dml = False
+            # cls_use_dml = False
+            # rec_use_dml = False
+
+            # # If we set everything to true onnx-runtime would automatically choose the fastest accelerator
+            # if self.options.device == self.options.Device.AUTO:
+            #     cls_use_cuda = True
+            #     rec_use_cuda = True
+            #     det_use_cuda = True
+            #     det_use_dml = True
+            #     cls_use_dml = True
+            #     rec_use_dml = True
+
+            # # If we set use_cuda to true onnx would use the cuda device available in runtime if no cuda device is available it would run on CPU.
+            # elif self.options.device == self.options.Device.CUDA:
+            #     cls_use_cuda = True
+            #     rec_use_cuda = True
+            #     det_use_cuda = True
+
+            # # If we set use_dml to true onnx would use the dml device available in runtime if no dml device is available it would work on CPU.
+            # elif self.options.device == self.options.Device.DIRECTML:
+            #     det_use_dml = True
+            #     cls_use_dml = True
+            #     rec_use_dml = True
+
+            self.reader = RapidOCR(
+                text_score=self.options.text_score,
+                cls_use_cuda=cls_use_cuda,
+                rec_use_cuda=rec_use_cuda,
+                det_use_cuda=det_use_cuda,
+                det_use_dml=det_use_dml,
+                cls_use_dml=cls_use_dml,
+                rec_use_dml=rec_use_dml,
+                print_verbose=self.options.print_verbose,
+                det_model_path=self.options.det_model_path,
+                cls_model_path=self.options.cls_model_path,
+                rec_model_path=self.options.rec_model_path,
+            )
+
+    def __call__(
+        self, conv_res: ConversionResult, page_batch: Iterable[Page]
+    ) -> Iterable[Page]:
+
+        if not self.enabled:
+            yield from page_batch
+            return
+
+        for page in page_batch:
+
+            assert page._backend is not None
+            if not page._backend.is_valid():
+                yield page
+            else:
+                with TimeRecorder(conv_res, "ocr"):
+                    ocr_rects = self.get_ocr_rects(page)
+
+                    all_ocr_cells = []
+                    for ocr_rect in ocr_rects:
+                        # Skip zero area boxes
+                        if ocr_rect.area() == 0:
+                            continue
+                        high_res_image = page._backend.get_page_image(
+                            scale=self.scale, cropbox=ocr_rect
+                        )
+                        im = numpy.array(high_res_image)
+                        result, _ = self.reader(
+                            im,
+                            use_det=self.options.use_det,
+                            use_cls=self.options.use_cls,
+                            use_rec=self.options.use_rec,
+                        )
+
+                        del high_res_image
+                        del im
+
+                        cells = [
+                            OcrCell(
+                                id=ix,
+                                text=line[1],
+                                confidence=line[2],
+                                bbox=BoundingBox.from_tuple(
+                                    coord=(
+                                        (line[0][0][0] / self.scale) + ocr_rect.l,
+                                        (line[0][0][1] / self.scale) + ocr_rect.t,
+                                        (line[0][2][0] / self.scale) + ocr_rect.l,
+                                        (line[0][2][1] / self.scale) + ocr_rect.t,
+                                    ),
+                                    origin=CoordOrigin.TOPLEFT,
+                                ),
+                            )
+                            for ix, line in enumerate(result)
+                        ]
+                        all_ocr_cells.extend(cells)
+
+                    # Post-process the cells
+                    page.cells = self.post_process_cells(all_ocr_cells, page.cells)
+
+                # DEBUG code:
+                if settings.debug.visualize_ocr:
+                    self.draw_ocr_rects_and_cells(conv_res, page, ocr_rects)
+
+                yield page

docling/pipeline/standard_pdf_pipeline.py

@@ -13,6 +13,7 @@ from docling.datamodel.pipeline_options import (
     EasyOcrOptions,
     OcrMacOptions,
     PdfPipelineOptions,
+    RapidOcrOptions,
     TesseractCliOcrOptions,
     TesseractOcrOptions,
 )
@@ -26,6 +27,7 @@ from docling.models.page_preprocessing_model import (
     PagePreprocessingModel,
     PagePreprocessingOptions,
 )
+from docling.models.rapid_ocr_model import RapidOcrModel
 from docling.models.table_structure_model import TableStructureModel
 from docling.models.tesseract_ocr_cli_model import TesseractOcrCliModel
 from docling.models.tesseract_ocr_model import TesseractOcrModel
@@ -121,6 +123,11 @@ class StandardPdfPipeline(PaginatedPipeline):
                 enabled=self.pipeline_options.do_ocr,
                 options=self.pipeline_options.ocr_options,
             )
+        elif isinstance(self.pipeline_options.ocr_options, RapidOcrOptions):
+            return RapidOcrModel(
+                enabled=self.pipeline_options.do_ocr,
+                options=self.pipeline_options.ocr_options,
+            )
         elif isinstance(self.pipeline_options.ocr_options, OcrMacOptions):
             if "darwin" != sys.platform:
                 raise RuntimeError(

{docling-2.7.1.dist-info → docling-2.8.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: docling
-Version: 2.7.1
+Version: 2.8.1
 Summary: SDK and CLI for parsing PDF, DOCX, HTML, and more, to a unified document representation for powering downstream workflows such as gen AI applications.
 Home-page: https://github.com/DS4SD/docling
 License: MIT
@@ -21,11 +21,12 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Provides-Extra: ocrmac
+Provides-Extra: rapidocr
 Provides-Extra: tesserocr
 Requires-Dist: beautifulsoup4 (>=4.12.3,<5.0.0)
 Requires-Dist: certifi (>=2024.7.4)
 Requires-Dist: deepsearch-glm (>=0.26.1,<0.27.0)
-Requires-Dist: docling-core (>=2.4.0,<3.0.0)
+Requires-Dist: docling-core (>=2.5.1,<3.0.0)
 Requires-Dist: docling-ibm-models (>=2.0.6,<3.0.0)
 Requires-Dist: docling-parse (>=2.0.5,<3.0.0)
 Requires-Dist: easyocr (>=1.7,<2.0)
@@ -34,14 +35,16 @@ Requires-Dist: huggingface_hub (>=0.23,<1)
 Requires-Dist: lxml (>=4.0.0,<6.0.0)
 Requires-Dist: marko (>=2.1.2,<3.0.0)
 Requires-Dist: ocrmac (>=1.0.0,<2.0.0) ; (sys_platform == "darwin") and (extra == "ocrmac")
+Requires-Dist: onnxruntime (>=1.7.0,<1.20.0) ; (python_version < "3.10") and (extra == "rapidocr")
+Requires-Dist: onnxruntime (>=1.7.0,<2.0.0) ; (python_version >= "3.10") and (extra == "rapidocr")
 Requires-Dist: openpyxl (>=3.1.5,<4.0.0)
 Requires-Dist: pandas (>=2.1.4,<3.0.0)
-Requires-Dist: pyarrow (>=16.1.0,<17.0.0)
 Requires-Dist: pydantic (>=2.0.0,<2.10)
 Requires-Dist: pydantic-settings (>=2.3.0,<3.0.0)
 Requires-Dist: pypdfium2 (>=4.30.0,<5.0.0)
 Requires-Dist: python-docx (>=1.1.2,<2.0.0)
 Requires-Dist: python-pptx (>=1.0.2,<2.0.0)
+Requires-Dist: rapidocr-onnxruntime (>=1.4.0,<2.0.0) ; (python_version < "3.13") and (extra == "rapidocr")
 Requires-Dist: requests (>=2.32.3,<3.0.0)
 Requires-Dist: rtree (>=1.3.0,<2.0.0)
 Requires-Dist: scipy (>=1.6.0,<2.0.0)
@@ -56,7 +59,7 @@ Description-Content-Type: text/markdown
   </a>
 </p>
 
-# Docling
+# 🦆 Docling
 
 <p align="center">
   <a href="https://trendshift.io/repositories/12132" target="_blank"><img src="https://trendshift.io/api/badge/repositories/12132" alt="DS4SD%2Fdocling | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a>
@@ -81,7 +84,7 @@ Docling parses documents and exports them to the desired format with ease and sp
 * 🗂️ Reads popular document formats (PDF, DOCX, PPTX, XLSX, Images, HTML, AsciiDoc & Markdown) and exports to Markdown and JSON
 * 📑 Advanced PDF document understanding including page layout, reading order & table structures
 * 🧩 Unified, expressive [DoclingDocument](https://ds4sd.github.io/docling/concepts/docling_document/) representation format
-* 🤖 Easy integration with LlamaIndex 🦙 & LangChain 🦜🔗 for powerful RAG / QA applications
+* 🤖 Easy integration with 🦙 LlamaIndex & 🦜🔗 LangChain for powerful RAG / QA applications
 * 🔍 OCR support for scanned PDFs
 * 💻 Simple and convenient CLI
 
@@ -117,8 +120,24 @@ result = converter.convert(source)
 print(result.document.export_to_markdown())  # output: "## Docling Technical Report[...]"
 ```
 
-Check out [Getting started](https://ds4sd.github.io/docling/).
-You will find lots of tuning options to leverage all the advanced capabilities.
+More [advanced usage options](https://ds4sd.github.io/docling/usage/) are available in
+the docs.
+
+## Documentation
+
+Check out Docling's [documentation](https://ds4sd.github.io/docling/), for details on
+installation, usage, concepts, recipes, extensions, and more.
+
+## Examples
+
+Go hands-on with our [examples](https://ds4sd.github.io/docling/examples/),
+demonstrating how to address different application use cases with Docling.
+
+## Integrations
+
+To further accelerate your AI application development, check out Docling's native
+[integrations](https://ds4sd.github.io/docling/integrations/) with popular frameworks
+and tools.
 
 ## Get help and support
 

{docling-2.7.1.dist-info → docling-2.8.1.dist-info}/RECORD

@@ -8,15 +8,15 @@ docling/backend/html_backend.py,sha256=qbu1W8xoTGnXMuZPRPLq68hDbCEj6ygnpxP5gYaod
 docling/backend/md_backend.py,sha256=tmuSCghjor9PqKIiVieCuZ4_t5JEjZMy3cq7u3yTgyU,14032
 docling/backend/msexcel_backend.py,sha256=23qUEScqr5GhY06xiqg-eBQ_JlAqO0FkPEmX6554sVA,12040
 docling/backend/mspowerpoint_backend.py,sha256=QD0NaatTO8U9CIFoiipkq3X5HxLZaaahH8nlrQ6ecDA,15710
-docling/backend/msword_backend.py,sha256=sMumfB9Xa2Md1a8WO-fGPPAKf1s3mCvErMyZ-xnBC2E,18495
+docling/backend/msword_backend.py,sha256=VFHPr-gCak7w3NJToc5Cs-JaTb4Vm3a1JnnRIfJO3TI,18526
 docling/backend/pdf_backend.py,sha256=unnw7QiRE1VXg6Pj-eYrtnFGrp5SSYiI324OlFxyv6c,2050
 docling/backend/pypdfium2_backend.py,sha256=B4bfv-dfzlWiKTfF8LN5fto_99YBu8A2c1_XIVwRUWI,8996
 docling/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-docling/cli/main.py,sha256=MpjbAXhOlbGnAnl5_OaKCdub61YPQBy1NOqroXQtNYE,10722
+docling/cli/main.py,sha256=AgPD32NfM0_bmHeKjx5-fqk57ahX5tN3AeoDOerhTuE,11808
 docling/datamodel/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/datamodel/base_models.py,sha256=6qlwPamDZ3XUsE2kTAyGKG6O2IJClVjCqaE7DZ74KHU,5533
 docling/datamodel/document.py,sha256=9dQf_J18X_MEWs-Mg3Ed6BykFPJ79ETmkkxcssY-vYo,20698
-docling/datamodel/pipeline_options.py,sha256=aC_CmtEhNLIbn9n3JuYhL_aA8UA0vFgw7HcGMUuOI4o,3117
+docling/datamodel/pipeline_options.py,sha256=J-6kWugUrxahymKzgaEgiqPuyle1fbInPXV2wNos6Vc,4550
 docling/datamodel/settings.py,sha256=JK8lZPBjUx2kD2q-Qpg-o3vOElADMcyQbRUL0EHZ7us,1263
 docling/document_converter.py,sha256=L0A3g7IQBaKIK7dWpUFC72ZqKywIPYkyh71Qd6DiNPE,10940
 docling/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -28,20 +28,21 @@ docling/models/layout_model.py,sha256=ZvbTSyxvXB5yLHNEti0Wv3trz0vwGuHySI5TCdApb0
 docling/models/ocr_mac_model.py,sha256=bLP14UUmZcSzjDe-HLj-mtksTuBmsCTg2C1wCxUpan0,4502
 docling/models/page_assemble_model.py,sha256=kSGNiRKhmzkpFH7xCiT3rulMsgJmUXFa6Th_eB-cLEk,7103
 docling/models/page_preprocessing_model.py,sha256=1gVrZjObKxAvXkkKvXnIFApPOggzgiTFPtt1CGbMbSs,2763
+docling/models/rapid_ocr_model.py,sha256=VQ0jaFmOzB9f-1JaqZ6d0o_El55Lr-nsFHfTNubMAuc,6005
 docling/models/table_structure_model.py,sha256=-ANSQpiN2avt3B9sbi7dHcoULUJbMBalAR5xxlrM7To,8421
 docling/models/tesseract_ocr_cli_model.py,sha256=OfopQnt2FGwtLJTMtW9jbJZ9EN2G2QFkA_aACjuUuDs,6372
 docling/models/tesseract_ocr_model.py,sha256=RDf6iV1q-oXaGfZXv0bW6SqjHNKQvBUDlUsOkuz0neY,6095
 docling/pipeline/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/pipeline/base_pipeline.py,sha256=IF1XWYgUGbdB4-teLkmM4Hvg_UNEfPrGuhExMRTUsk8,7168
 docling/pipeline/simple_pipeline.py,sha256=mZqANqUtAOFAyqQEShErQnAUz6tJFOl6zVoazEDJ_wE,2254
-docling/pipeline/standard_pdf_pipeline.py,sha256=btm_y1ZsjUrtWvMbF6RA8BVM0ENrK4z_rqF0jjdeZmU,8473
+docling/pipeline/standard_pdf_pipeline.py,sha256=7sbkh9EwXlhSfJSgf-WyjB5jdJ1El7Pn4siSssTJpq8,8789
 docling/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 docling/utils/export.py,sha256=KyGF1BVDHPFfHVXZc8vegsWlFfOgGPP2YckWpTadyI8,4694
 docling/utils/layout_utils.py,sha256=vlN0rc8i0ayRGn3WnaG-pdmqEL00KKGl2zez3Gj-hrk,32074
 docling/utils/profiling.py,sha256=YaMGoB9MMZpagF9mb5ndoHj8Lpb9aIdb7El-Pl7IcFs,1753
 docling/utils/utils.py,sha256=llhXSbIDNZ1MHOwBEfLHBAoJIAYI7QlPIonlI1jLUJ0,1208
-docling-2.7.1.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
-docling-2.7.1.dist-info/METADATA,sha256=TvD3BGlbO1ci54NzwmLxqSITXIdMefyj71YjdZkD7Vs,6906
-docling-2.7.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-docling-2.7.1.dist-info/entry_points.txt,sha256=VOSzV77znM52dz5ysaDuJ0ijl1cnfrh1ZPg8od5OcTs,48
-docling-2.7.1.dist-info/RECORD,,
+docling-2.8.1.dist-info/LICENSE,sha256=mBb7ErEcM8VS9OhiGHnQ2kk75HwPhr54W1Oiz3965MY,1088
+docling-2.8.1.dist-info/METADATA,sha256=auj5PtDj-UBB72sW8jk1CSVSwQpd9q0nYzoAYIItl8o,7682
+docling-2.8.1.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+docling-2.8.1.dist-info/entry_points.txt,sha256=VOSzV77znM52dz5ysaDuJ0ijl1cnfrh1ZPg8od5OcTs,48
+docling-2.8.1.dist-info/RECORD,,