Compare requested languages to OCR engine instead of tesseract directly

Also refactoring to facilitating validation needing the plugin manager.

src/ocrmypdf/__main__.py

@@ -58,7 +58,7 @@ def run(args=None):
     )
     log.debug('ocrmypdf %s', __version__)
     try:
-        check_options(options)
+        check_options(options, plugin_manager)
     except ValueError as e:
         log.error(e)
         return ExitCode.bad_args

src/ocrmypdf/_validation.py

@@ -27,6 +27,7 @@ from shutil import copyfileobj
 
 import PIL
 
+from ocrmypdf._plugin_manager import get_plugin_manager
 from ocrmypdf._unicodefun import verify_python3_env
 from ocrmypdf.exceptions import (
     BadArgsError,
@@ -72,7 +73,7 @@ def check_platform():
         )
 
 
-def check_options_languages(options):
+def check_options_languages(options, plugin_manager):
     if not options.language:
         options.language = [DEFAULT_LANGUAGE]
         system_lang = locale.getlocale()[0]
@@ -84,12 +85,13 @@ def check_options_languages(options):
         options.language = options.language[0].split('+')
 
     languages = set(options.language)
-    if not languages.issubset(tesseract.get_languages(options.tesseract_env)):
+    ocr_engine = plugin_manager.hook.get_ocr_engine()
+    if not languages.issubset(ocr_engine.languages(options)):
         msg = (
-            "The installed version of tesseract does not have language "
-            "data for the following requested languages: \n"
+            f"{ocr_engine} does not have language data for the following "
+            "requested languages: \n"
         )
-        for lang in languages - tesseract.get_languages(options.tesseract_env):
+        for lang in languages - ocr_engine.languages(options):
             msg += lang + '\n'
         raise MissingDependencyError(msg)
 
@@ -308,9 +310,9 @@ def check_options_pillow(options):
         PIL.Image.MAX_IMAGE_PIXELS = None
 
 
-def check_options(options):
+def check_options(options, plugin_manager):
     check_platform()
-    check_options_languages(options)
+    check_options_languages(options, plugin_manager)
     check_options_metadata(options)
     check_options_output(options)
     check_options_sidecar(options)

src/ocrmypdf/api.py

@@ -279,5 +279,5 @@ def ocr(  # pylint: disable=unused-argument
     options = create_options(
         **{k: v for k, v in locals().items() if not k.startswith('_')}
     )
-    check_options(options)
+    check_options(options, _plugin_manager)
     return run_pipeline(options=options, plugin_manager=_plugin_manager, api=True)

src/ocrmypdf/builtin_plugins/tesseract_ocr.py

@@ -30,9 +30,12 @@ class TesseractOcrEngine(OcrEngine):
         tag = '-PDF' if options.pdf_renderer == 'sandwich' else ''
         return f"Tesseract OCR{tag} {TesseractOcrEngine.version()}"
 
+    def __str__(self):
+        return f"Tesseract OCR {TesseractOcrEngine.version()}"
+
     @staticmethod
-    def languages():
-        return tesseract.get_languages()
+    def languages(options):
+        return tesseract.get_languages(options.tesseract_env)
 
     @staticmethod
     def get_orientation(input_file, options):

src/ocrmypdf/pluginspec.py

@@ -100,11 +100,15 @@ class OcrEngine(ABC):
         """Returns the version of the OCR engine."""
 
     @abstractstaticmethod
-    def creator_tag(options) -> str:
+    def creator_tag(options: Namespace) -> str:
         """Returns the creator tag to identify this software's role in creating the PDF."""
 
     @abstractstaticmethod
-    def languages() -> AbstractSet[str]:
+    def __str__(self):
+        """Returns name of OCR engine and version."""
+
+    @abstractstaticmethod
+    def languages(options: Namespace) -> AbstractSet[str]:
         """Returns set of languages that are supported."""
 
     @abstractstaticmethod

tests/conftest.py

@@ -25,6 +25,7 @@ from subprocess import PIPE, run
 import pytest
 
 from ocrmypdf import api, cli, pdfinfo
+from ocrmypdf._plugin_manager import get_plugin_manager
 from ocrmypdf.exec import unpaper
 
 pytest_plugins = ['helpers_namespace']
@@ -217,11 +218,12 @@ def check_ocrmypdf(input_file, output_file, *args, env=None):
         [str(input_file), str(output_file)]
         + [str(arg) for arg in args if arg is not None]
     )
-    api.check_options(options)
+    plugin_manager = get_plugin_manager(options.plugins)
+    api.check_options(options, plugin_manager)
     if env:
         options.tesseract_env = env
         options.tesseract_env['_OCRMYPDF_TEST_INFILE'] = os.fspath(input_file)
-    result = api.run_pipeline(options, plugin_manager=None, api=True)
+    result = api.run_pipeline(options, plugin_manager=plugin_manager, api=True)
 
     assert result == 0
     assert output_file.exists(), "Output file not created"
@@ -251,7 +253,8 @@ def run_ocrmypdf_api(input_file, output_file, *args, env=None):
     if options.tesseract_env:
         assert all(isinstance(v, (str, bytes)) for v in options.tesseract_env.values())
 
-    api.check_options(options)
+    plugin_manager = get_plugin_manager(options.plugins)
+    api.check_options(options, plugin_manager)
     return api.run_pipeline(options, plugin_manager=None, api=False)
 
 

tests/test_unpaper.py

@@ -20,6 +20,7 @@ from unittest.mock import patch
 
 import pytest
 
+from ocrmypdf._plugin_manager import get_plugin_manager
 from ocrmypdf._validation import check_options
 from ocrmypdf.cli import get_parser
 from ocrmypdf.exceptions import ExitCode, MissingDependencyError
@@ -43,11 +44,12 @@ def test_no_unpaper(resources, no_outpdf):
     input_ = fspath(resources / "c02-22.pdf")
     output = fspath(no_outpdf)
     options = get_parser().parse_args(args=["--clean", input_, output])
-
+    plugin_manager = get_plugin_manager(options.plugins)
     with patch("ocrmypdf.exec.unpaper.version") as mock_unpaper_version:
         mock_unpaper_version.side_effect = FileNotFoundError("unpaper")
+
         with pytest.raises(MissingDependencyError):
-            check_options(options)
+            check_options(options, plugin_manager)
 
 
 def test_old_unpaper(spoof_unpaper_oldversion, resources, no_outpdf):

tests/test_validation.py

@@ -22,6 +22,7 @@ from unittest.mock import patch
 import pytest
 
 import ocrmypdf._validation as vd
+from ocrmypdf._plugin_manager import get_plugin_manager
 from ocrmypdf.api import create_options
 from ocrmypdf.cli import get_parser
 from ocrmypdf.exceptions import BadArgsError, MissingDependencyError
@@ -153,8 +154,9 @@ def test_false_action_store_true():
 @pytest.mark.parametrize('progress_bar', [True, False])
 def test_no_progress_bar(progress_bar, resources):
     opts = make_opts(progress_bar=progress_bar, input_file=(resources / 'trivial.pdf'))
+    plugin_manager = get_plugin_manager(opts.plugins)
     with patch('ocrmypdf._concurrent.tqdm', autospec=True) as tqdmpatch:
-        vd.check_options(opts)
+        vd.check_options(opts, plugin_manager)
         pdfinfo = PdfInfo(opts.input_file, progbar=opts.progress_bar)
         assert pdfinfo is not None
         assert tqdmpatch.called
@@ -164,11 +166,12 @@ def test_no_progress_bar(progress_bar, resources):
 
 def test_language_warning(caplog):
     opts = make_opts(language=None)
+    plugin_manager = get_plugin_manager(opts.plugins)
     caplog.set_level(logging.DEBUG)
     with patch(
         'ocrmypdf._validation.locale.getlocale', return_value=('en_US', 'UTF-8')
     ):
-        vd.check_options_languages(opts)
+        vd.check_options_languages(opts, plugin_manager)
         assert opts.language == ['eng']
         assert '' in caplog.text
 
@@ -176,7 +179,7 @@ def test_language_warning(caplog):
     with patch(
         'ocrmypdf._validation.locale.getlocale', return_value=('fr_FR', 'UTF-8')
     ):
-        vd.check_options_languages(opts)
+        vd.check_options_languages(opts, plugin_manager)
         assert opts.language == ['eng']
         assert 'assuming --language' in caplog.text