feat: add optional ocrmac support

- Added `ocrmac` as an optional dependency in `pyproject.toml` and `poetry.lock`.
- Updated the `[tool.poetry.extras]` section to include `ocrmac`.
- Modified end-to-end OCR conversion tests to support `OcrMacOptions` on macOS.

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

poetry.lock

@@ -7360,9 +7360,10 @@ test = ["big-O", "importlib-resources", "jaraco.functools", "jaraco.itertools",
 type = ["pytest-mypy"]
 
 [extras]
+ocrmac = ["ocrmac"]
 tesserocr = ["tesserocr"]
 
 [metadata]
 lock-version = "2.0"
 python-versions = "^3.10"
-content-hash = "22ce94f59bd5912d1d43d3583318fe4672ee1c97a35d5dbec46fb0a222ea09ff"
+content-hash = "a9912d5333298b629d774f669f4e5eb794333d1694ca7f72f56b9a37e51d5776"

pyproject.toml

@@ -48,6 +48,7 @@ beautifulsoup4 = "^4.12.3"
 pandas = "^2.1.4"
 marko = "^2.1.2"
 openpyxl = "^3.1.5"
+ocrmac = { version = "^1.0.0", optional = true }
 
 [tool.poetry.group.dev.dependencies]
 black = {extras = ["jupyter"], version = "^24.4.2"}
@@ -92,10 +93,10 @@ torchvision = [
   {markers = "sys_platform != 'darwin' or platform_machine != 'x86_64'", version = "^0"},
   {markers = "sys_platform == 'darwin' and platform_machine == 'x86_64'", version = "~0.17.2"}
 ]
-ocrmac = {markers = "sys_platform == 'darwin'", version = "^1.0.0", optional = true}
 
 [tool.poetry.extras]
 tesserocr = ["tesserocr"]
+ocrmac = ["ocrmac"]
 
 [tool.poetry.scripts]
 docling = "docling.cli.main:app"

tests/test_e2e_ocr_conversion.py

@@ -63,6 +63,7 @@ def test_e2e_conversions():
 
     # only works on mac
     if "darwin" == sys.platform:
+        engines.append(OcrMacOptions())
         engines.append(OcrMacOptions(force_full_page_ocr=True))
 
     for ocr_options in engines: