Refactor --language argument into set

src/ocrmypdf/_validation.py

@@ -74,24 +74,19 @@ def check_platform():
 
 
 def check_options_languages(options, plugin_manager):
-    if not options.language:
-        options.language = [DEFAULT_LANGUAGE]
+    if not options.languages:
+        options.languages = {DEFAULT_LANGUAGE}
         system_lang = locale.getlocale()[0]
         if system_lang and not system_lang.startswith('en'):
             log.debug("No language specified; assuming --language %s", DEFAULT_LANGUAGE)
 
-    # Support v2.x "eng+deu" language syntax
-    if '+' in options.language[0]:
-        options.language = options.language[0].split('+')
-
-    languages = set(options.language)
     ocr_engine = plugin_manager.hook.get_ocr_engine()
-    if not languages.issubset(ocr_engine.languages(options)):
+    if not options.languages.issubset(ocr_engine.languages(options)):
         msg = (
             f"{ocr_engine} does not have language data for the following "
             "requested languages: \n"
         )
-        for lang in languages - ocr_engine.languages(options):
+        for lang in options.languages - ocr_engine.languages(options):
             msg += lang + '\n'
         raise MissingDependencyError(msg)
 
@@ -101,8 +96,7 @@ def check_options_output(options):
     # 1. Ghostscript < 9.20 mangles multibyte Unicode
     # 2. hocr doesn't work on non-Latin languages (so don't select it)
 
-    languages = set(options.language)
-    is_latin = languages.issubset(HOCR_OK_LANGS)
+    is_latin = options.languages.issubset(HOCR_OK_LANGS)
 
     if options.pdf_renderer == 'hocr' and not is_latin:
         msg = (

src/ocrmypdf/builtin_plugins/tesseract_ocr.py

@@ -95,7 +95,7 @@ def check_options(options):
         options.pdf_renderer = 'sandwich'
 
     if options.pdf_renderer == 'sandwich' and not tesseract.has_textonly_pdf(
-        options.tesseract_env, set(options.language)
+        options.tesseract_env, set(options.languages)
     ):
         raise MissingDependencyError(
             "You are using an alpha version of Tesseract 4.0 that does not support "
@@ -147,7 +147,7 @@ class TesseractOcrEngine(OcrEngine):
             input_file=input_file,
             output_hocr=output_hocr,
             output_text=output_text,
-            languages=options.language,
+            languages=options.languages,
             engine_mode=options.tesseract_oem,
             tessconfig=options.tesseract_config,
             timeout=options.tesseract_timeout,
@@ -163,7 +163,7 @@ class TesseractOcrEngine(OcrEngine):
             input_file=input_file,
             output_pdf=output_pdf,
             output_text=output_text,
-            languages=options.language,
+            languages=options.languages,
             engine_mode=options.tesseract_oem,
             tessconfig=options.tesseract_config,
             timeout=options.tesseract_timeout,

src/ocrmypdf/cli.py

@@ -54,6 +54,20 @@ class ArgumentParser(argparse.ArgumentParser):
         raise ValueError(message)
 
 
+class LanguageSetAction(argparse.Action):
+    def __init__(self, option_strings, dest, default=None, **kwargs):
+        if default is None:
+            default = set()
+        super().__init__(option_strings, dest, default=default, **kwargs)
+
+    def __call__(self, parser, namespace, values, option_string=None):
+        dest = getattr(namespace, self.dest)
+        if '+' in values:
+            dest.add(lang for lang in values.split('+'))
+        else:
+            dest.add(values)
+
+
 def get_parser():
     parser = ArgumentParser(
         prog=_PROGRAM_NAME,
@@ -126,7 +140,8 @@ Online documentation is located at:
     parser.add_argument(
         '-l',
         '--language',
-        action='append',
+        dest='languages',
+        action=LanguageSetAction,
         help="Language(s) of the file to be OCRed (see tesseract --list-langs for "
         "all language packs installed in your system). Use -l eng+deu for "
         "multiple languages.",

tests/test_validation.py

@@ -181,7 +181,7 @@ def test_language_warning(caplog):
         'ocrmypdf._validation.locale.getlocale', return_value=('en_US', 'UTF-8')
     ):
         vd.check_options_languages(opts, plugin_manager)
-        assert opts.language == ['eng']
+        assert opts.languages == {'eng'}
         assert '' in caplog.text
 
     opts = make_opts(language=None)
@@ -189,7 +189,7 @@ def test_language_warning(caplog):
         'ocrmypdf._validation.locale.getlocale', return_value=('fr_FR', 'UTF-8')
     ):
         vd.check_options_languages(opts, plugin_manager)
-        assert opts.language == ['eng']
+        assert opts.languages == {'eng'}
         assert 'assuming --language' in caplog.text