feat: add support for ocrmac OCR engine on macOS (#276)

* feat: add support for `ocrmac` OCR engine on macOS

- Integrates `ocrmac` as an OCR engine option for macOS users.
- Adds configuration options and dependencies for `ocrmac`.
- Updates documentation to reflect new engine support.

This change allows macOS users to utilize `ocrmac` for improved OCR performance and compatibility.

Signed-off-by: Suhwan Seo <nuridol@gmail.com>

* updated the poetry lock

Signed-off-by: Suhwan Seo <nuridol@gmail.com>

* Fix linting issues, update CLI docs, and add error for ocrmac use on non-Mac systems

- Resolved formatting and linting issues
- Updated `--ocr-engine` CLI option documentation for `ocrmac`
- Added RuntimeError for attempts to use `ocrmac` on non-Mac platforms

Signed-off-by: Suhwan Seo <nuridol@gmail.com>

* feat: add support for `ocrmac` OCR engine on macOS

- Integrates `ocrmac` as an OCR engine option for macOS users.
- Adds configuration options and dependencies for `ocrmac`.
- Updates documentation to reflect new engine support.

This change allows macOS users to utilize `ocrmac` for improved OCR performance and compatibility.

Signed-off-by: Suhwan Seo <nuridol@gmail.com>

* docs: update examples and installation for ocrmac support

- Added `OcrMacOptions` to `custom_convert.py` and `full_page_ocr.py` examples.
- Included usage comments and examples for `OcrMacOptions` in OCR pipelines.
- Updated installation guide to include instructions for installing `ocrmac`, noting macOS version requirements (10.15+).
- Highlighted that `ocrmac` leverages Apple's Vision framework as an OCR backend.

This enhances documentation for users working on macOS to leverage `ocrmac` effectively.

Signed-off-by: Suhwan Seo <nuridol@gmail.com>

* fix: update `ocrmac` dependency with macOS-specific marker

- Added `sys_platform == 'darwin'` marker to the `ocrmac` dependency in `pyproject.toml` to specify macOS compatibility.
- Updated the content hash in `poetry.lock` to reflect the changes.

This ensures the `ocrmac` dependency is only installed on macOS systems.

Signed-off-by: Suhwan Seo <nuridol@gmail.com>

---------

Signed-off-by: Suhwan Seo <nuridol@gmail.com>
Co-authored-by: Suhwan Seo <nuridol@gmail.com>

docs/examples/custom_convert.py

@@ -7,6 +7,7 @@ from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend
 from docling.datamodel.base_models import InputFormat
 from docling.datamodel.pipeline_options import PdfPipelineOptions
 from docling.document_converter import DocumentConverter, PdfFormatOption
+from docling.models.ocr_mac_model import OcrMacOptions
 from docling.models.tesseract_ocr_cli_model import TesseractCliOcrOptions
 from docling.models.tesseract_ocr_model import TesseractOcrOptions
 
@@ -122,6 +123,20 @@ def main():
     #     }
     # )
 
+    # Docling Parse with ocrmac(Mac only)
+    # ----------------------
+    # pipeline_options = PdfPipelineOptions()
+    # pipeline_options.do_ocr = True
+    # pipeline_options.do_table_structure = True
+    # pipeline_options.table_structure_options.do_cell_matching = True
+    # pipeline_options.ocr_options = OcrMacOptions()
+
+    # doc_converter = DocumentConverter(
+    #     format_options={
+    #         InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)
+    #     }
+    # )
+
     ###########################################################################
 
     start_time = time.time()

docs/examples/full_page_ocr.py

@@ -4,6 +4,7 @@ from docling.backend.docling_parse_backend import DoclingParseDocumentBackend
 from docling.datamodel.base_models import InputFormat
 from docling.datamodel.pipeline_options import (
     EasyOcrOptions,
+    OcrMacOptions,
     PdfPipelineOptions,
     TesseractCliOcrOptions,
     TesseractOcrOptions,
@@ -19,9 +20,10 @@ def main():
     pipeline_options.do_table_structure = True
     pipeline_options.table_structure_options.do_cell_matching = True
 
-    # Any of the OCR options can be used:EasyOcrOptions, TesseractOcrOptions, TesseractCliOcrOptions
+    # Any of the OCR options can be used:EasyOcrOptions, TesseractOcrOptions, TesseractCliOcrOptions, OcrMacOptions(Mac only)
     # ocr_options = EasyOcrOptions(force_full_page_ocr=True)
     # ocr_options = TesseractOcrOptions(force_full_page_ocr=True)
+    # ocr_options = OcrMacOptions(force_full_page_ocr=True)
     ocr_options = TesseractCliOcrOptions(force_full_page_ocr=True)
     pipeline_options.ocr_options = ocr_options
 

docs/installation.md

@@ -30,6 +30,7 @@ Works on macOS, Linux, and Windows, with support for both x86_64 and arm64 archi
     | [EasyOCR](https://github.com/JaidedAI/EasyOCR) | Default in Docling or via `pip install easyocr`. | `EasyOcrOptions` |
     | Tesseract | System dependency. See description for Tesseract and Tesserocr below.  | `TesseractOcrOptions` |
     | Tesseract CLI | System dependency. See description below. | `TesseractCliOcrOptions` |
+    | OcrMac | System dependency. See description below. | `OcrMacOptions` |
 
     The Docling `DocumentConverter` allows to choose the OCR engine with the `ocr_options` settings. For example
 
@@ -91,6 +92,17 @@ Works on macOS, Linux, and Windows, with support for both x86_64 and arm64 archi
     pip install --no-binary :all: tesserocr
     ```
 
+    <h3>ocrmac installation</h3>
+
+    [ocrmac](https://github.com/straussmaximilian/ocrmac) is using
+    Apple's vision(or livetext) framework as OCR backend.
+    For using this engine with Docling, ocrmac must be installed on your system.
+    This only works on macOS systems with newer macOS versions (10.15+).
+
+    ```console
+    pip install ocrmac
+    ```
+
 ## Development setup
 
 To develop Docling features, bugfixes etc., install as follows from your local clone's root dir: