from pathlib import Path import sys import pytest sys.path.insert(0, str(Path(__file__).resolve().parents[1])) from modules.asr import subtitle_recognition def test_recognize_audio_subtitles_dispatches_qwen3_asr(monkeypatch, tmp_path): audio_path = tmp_path / "audio.wav" asr_path = tmp_path / "qwen" aligner_path = tmp_path / "aligner" calls = {} monkeypatch.setattr( subtitle_recognition, "require_qwen_asr_model_dirs", lambda: (asr_path, aligner_path), ) monkeypatch.setattr(subtitle_recognition.qwen_asr, "missing_dependencies", lambda: []) def fake_recognize(*args): calls["recognize"] = args return [] monkeypatch.setattr( subtitle_recognition.qwen_asr, "recognize_subtitle_segments", fake_recognize, ) monkeypatch.setattr( subtitle_recognition, "smart_split_subtitle_segments", lambda segments, maximum: calls.setdefault("split", (segments, maximum)) and segments, raising=False, ) subtitle_recognition.recognize_audio_subtitles( audio_path, "qwen3-asr", max_sentence_length=24, unload_model=False, ) assert calls["recognize"] == (audio_path, asr_path, aligner_path, False) assert calls["split"] == ([], 24) def test_recognize_audio_subtitles_dispatches_whisper(monkeypatch, tmp_path): audio_path = tmp_path / "audio.wav" model_path = tmp_path / "whisper.safetensors" calls = {} monkeypatch.setattr( subtitle_recognition, "require_whisper_large_v3_model_path", lambda: model_path, ) monkeypatch.setattr(subtitle_recognition.whisper_asr, "missing_dependencies", lambda: []) def fake_recognize(*args): calls["recognize"] = args return [] monkeypatch.setattr( subtitle_recognition.whisper_asr, "recognize_subtitle_segments", fake_recognize, ) monkeypatch.setattr( subtitle_recognition, "smart_split_subtitle_segments", lambda segments, maximum: calls.setdefault("split", (segments, maximum)) and segments, raising=False, ) subtitle_recognition.recognize_audio_subtitles( audio_path, "whisper-large-v3", max_sentence_length=32, unload_model=True, ) assert calls["recognize"] == (audio_path, model_path, True) assert calls["split"] == ([], 32) def test_subtitle_recognition_options_validate_api_settings(): assert subtitle_recognition.subtitle_recognition_options({}) == (20, True) assert subtitle_recognition.subtitle_recognition_options({ "max_sentence_length": 32, "unload_model": False, }) == (32, False) with pytest.raises(ValueError, match="max_sentence_length"): subtitle_recognition.subtitle_recognition_options({"max_sentence_length": 0}) with pytest.raises(ValueError, match="unload_model"): subtitle_recognition.subtitle_recognition_options({"unload_model": "false"}) def test_recognize_audio_subtitles_reports_missing_dependencies(monkeypatch, tmp_path): monkeypatch.setattr( subtitle_recognition, "require_whisper_large_v3_model_path", lambda: tmp_path / "whisper.safetensors", ) monkeypatch.setattr( subtitle_recognition.whisper_asr, "missing_dependencies", lambda: ["openai-whisper"], ) with pytest.raises( subtitle_recognition.MissingSubtitleRecognitionDependenciesError ) as error: subtitle_recognition.recognize_audio_subtitles( tmp_path / "audio.wav", "whisper-large-v3", ) assert error.value.dependencies == ["openai-whisper"] def test_recognize_audio_subtitles_rejects_unknown_model(tmp_path): with pytest.raises(ValueError, match="model_type"): subtitle_recognition.recognize_audio_subtitles(tmp_path / "audio.wav", "unknown")