import unittest

from core.pronunciations import (
    expand_abbreviations,
    normalize_ocr_text,
    prepare_text_for_tts,
    remove_linebreak_hyphens,
)


class PronunciationTests(unittest.TestCase):
    def test_fuses_printed_line_break_but_keeps_real_compound(self):
        self.assertEqual(remove_linebreak_hyphens("program-\nmieren mit einem KI-Modell"),
                         "programmieren mit einem KI-Modell")
        self.assertEqual(remove_linebreak_hyphens("Hardware- und Software"),
                         "Hardware- und Software")
        self.assertEqual(remove_linebreak_hyphens("einge\u00adtrichtert"), "eingetrichtert")

    def test_existing_flattened_ocr_layouts_are_repaired_too(self):
        self.assertEqual(remove_linebreak_hyphens("eingetrich- tert"), "eingetrichtert")

    def test_german_umlaut_and_acronyms_are_prepared_for_speech(self):
        spoken = prepare_text_for_tts(
            "KI wird jahrlich wichtiger und APIs sind naturlich verfügbar.", "de")
        self.assertIn("Künstliche Intelligenz", spoken)
        self.assertIn("jährlich", spoken)
        self.assertIn("Programmierschnittstellen", spoken)
        self.assertIn("natürlich", spoken)

    def test_english_acronyms_use_english_expansions(self):
        self.assertEqual(expand_abbreviations("AI uses an API.", "en"),
                         "artificial intelligence uses an application programming interface.")

    def test_unknown_technical_words_and_names_stay_unchanged(self):
        text = "OpenAI und PyMuPDF bleiben unverändert."
        self.assertEqual(normalize_ocr_text(text, "de"), text)


if __name__ == "__main__":
    unittest.main()