Made max line length configurable

Author: beveradb

lyrics_transcriber/core/config.py

@@ -26,7 +26,7 @@ class OutputConfig:
     """Configuration for output generation."""
 
     output_styles_json: str
-    max_line_length: int = 36
+    default_max_line_length: int = 36
     styles: Dict[str, Any] = field(default_factory=dict)
     output_dir: Optional[str] = os.getcwd()
     cache_dir: str = os.getenv(

lyrics_transcriber/core/controller.py

@@ -109,6 +109,9 @@ def __init__(
         # Initialize results
         self.results = LyricsControllerResult()
 
+        # Load styles early so lyrics providers can use them
+        self._load_styles()
+
         # Initialize components (with dependency injection)
         self.transcribers = transcribers or self._initialize_transcribers()
         self.lyrics_providers = lyrics_providers or self._initialize_lyrics_providers()
@@ -127,6 +130,20 @@ def __init__(
         if self.output_config.render_video:
             self.logger.info(f"    Video resolution: {self.output_config.video_resolution}")
 
+    def _load_styles(self) -> None:
+        """Load styles from JSON file if available."""
+        if self.output_config.output_styles_json and os.path.exists(self.output_config.output_styles_json):
+            try:
+                with open(self.output_config.output_styles_json, "r") as f:
+                    self.output_config.styles = json.load(f)
+                self.logger.debug(f"Loaded output styles from: {self.output_config.output_styles_json}")
+            except Exception as e:
+                self.logger.warning(f"Failed to load output styles file: {str(e)}")
+                self.output_config.styles = {}
+        else:
+            self.logger.debug("No styles JSON file provided or file does not exist")
+            self.output_config.styles = {}
+
     def _sanitize_filename(self, filename: str) -> str:
         """Replace or remove characters that are unsafe for filenames."""
         if not filename:
@@ -189,6 +206,10 @@ def _initialize_lyrics_providers(self) -> Dict[str, BaseLyricsProvider]:
         """Initialize available lyrics providers."""
         providers = {}
 
+        # Get max_line_length from styles if available, otherwise use config default
+        max_line_length = self.output_config.styles.get("karaoke", {}).get("max_line_length", self.output_config.default_max_line_length)
+        self.logger.info(f"Using max_line_length for lyrics providers: {max_line_length}")
+
         # Create provider config with all necessary parameters
         provider_config = LyricsProviderConfig(
             genius_api_token=self.lyrics_config.genius_api_token,
@@ -197,6 +218,7 @@ def _initialize_lyrics_providers(self) -> Dict[str, BaseLyricsProvider]:
             lyrics_file=self.lyrics_config.lyrics_file,
             cache_dir=self.output_config.cache_dir,
             audio_filepath=self.audio_filepath,
+            max_line_length=max_line_length,
         )
 
         if provider_config.lyrics_file and os.path.exists(provider_config.lyrics_file):

lyrics_transcriber/lyrics/base_lyrics_provider.py

@@ -21,7 +21,7 @@ class LyricsProviderConfig:
     lyrics_file: Optional[str] = None
     cache_dir: Optional[str] = None
     audio_filepath: Optional[str] = None
-    max_line_length: int = 36  # New config parameter for KaraokeLyricsProcessor
+    max_line_length: int = 36  # Config parameter for KaraokeLyricsProcessor
 
 
 class BaseLyricsProvider(ABC):

lyrics_transcriber/output/generator.py

@@ -52,21 +52,31 @@ def __init__(
 
         self.logger.info(f"Initializing OutputGenerator with config: {self.config}")
 
-        if self.config.render_video or self.config.generate_cdg:
-            # Load output styles from JSON
+        # Load output styles from JSON if provided
+        if self.config.output_styles_json and os.path.exists(self.config.output_styles_json):
             try:
                 with open(self.config.output_styles_json, "r") as f:
                     self.config.styles = json.load(f)
                 self.logger.debug(f"Loaded output styles from: {self.config.output_styles_json}")
             except Exception as e:
-                raise ValueError(f"Failed to load output styles file: {str(e)}")
+                if self.config.render_video or self.config.generate_cdg:
+                    # Only raise error for video/CDG since they require styles
+                    raise ValueError(f"Failed to load output styles file: {str(e)}")
+                else:
+                    # For other outputs, just log warning and continue with empty styles
+                    self.logger.warning(f"Failed to load output styles file: {str(e)}")
+                    self.config.styles = {}
+        else:
+            # No styles file provided or doesn't exist
+            if self.config.render_video or self.config.generate_cdg:
+                raise ValueError(f"Output styles file required for video/CDG generation but not found: {self.config.output_styles_json}")
+            else:
+                self.config.styles = {}
 
         # Set video resolution parameters
         self.video_resolution_num, self.font_size, self.line_height = self._get_video_params(self.config.video_resolution)
         self.logger.info(f"Video resolution: {self.video_resolution_num}, font size: {self.font_size}, line height: {self.line_height}")
 
-        self.segment_resizer = SegmentResizer(max_line_length=self.config.max_line_length, logger=self.logger)
-
         # Initialize generators
         self.plain_text = PlainTextGenerator(self.config.output_dir, self.logger)
         self.lyrics_file = LyricsFileGenerator(self.config.output_dir, self.logger)
@@ -100,6 +110,12 @@ def __init__(
                         self.config.styles["karaoke"]["font_size"] = self.font_size
                         self.logger.info(f"Preview mode: Scaled down font_size to: {self.font_size}")
 
+        # Get max_line_length from styles if available, otherwise use config default
+        max_line_length = self.config.styles.get("karaoke", {}).get("max_line_length", self.config.default_max_line_length)
+        self.logger.info(f"Using max_line_length: {max_line_length}")
+        self.segment_resizer = SegmentResizer(max_line_length=max_line_length, logger=self.logger)
+
+        if self.config.render_video:
             # Initialize subtitle generator with potentially scaled values
             self.subtitle = SubtitlesGenerator(
                 output_dir=self.config.output_dir,

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "lyrics-transcriber"
-version = "0.67.0"
+version = "0.68.0"
 description = "Automatically create synchronised lyrics files in ASS and MidiCo LRC formats with word-level timestamps, using Whisper and lyrics from Genius and Spotify"
 authors = ["Andrew Beveridge <[email protected]>"]
 license = "MIT"

tests/test_helpers.py

@@ -353,7 +353,7 @@ def create_test_output_config(
         output_styles_json=output_styles_json,
         output_dir=output_dir,
         cache_dir=cache_dir,
-        max_line_length=36,
+        default_max_line_length=36,
         video_resolution="360p",
         render_video=render_video,
         generate_cdg=generate_cdg,

tests/unit/output/test_generator.py

@@ -36,7 +36,7 @@ def test_config_no_video(tmp_path, basic_styles_content):
         output_dir=str(tmp_path / "output"),
         cache_dir=str(tmp_path / "cache"),
         video_resolution="360p",
-        max_line_length=36,
+        default_max_line_length=36,
         render_video=False,  # Disable video to avoid complexity
         generate_cdg=False   # Disable CDG to avoid complexity
     )
@@ -54,7 +54,7 @@ def test_config_with_video(tmp_path, basic_styles_content):
         output_dir=str(tmp_path / "output"),
         cache_dir=str(tmp_path / "cache"),
         video_resolution="360p",
-        max_line_length=36,
+        default_max_line_length=36,
         render_video=True,
         generate_cdg=False
     )
@@ -88,10 +88,11 @@ def test_initialization_missing_styles_file(self, tmp_path):
             output_styles_json=str(tmp_path / "nonexistent.json"),
             output_dir=str(tmp_path / "output"),
             cache_dir=str(tmp_path / "cache"),
+            default_max_line_length=36,
             render_video=True
         )
 
-        with pytest.raises(ValueError, match="Failed to load output styles file"):
+        with pytest.raises(ValueError, match="Output styles file required for video/CDG generation but not found"):
             OutputGenerator(config=config)
 
     def test_initialization_invalid_styles_file(self, tmp_path):
@@ -104,6 +105,7 @@ def test_initialization_invalid_styles_file(self, tmp_path):
             output_styles_json=str(styles_path),
             output_dir=str(tmp_path / "output"),
             cache_dir=str(tmp_path / "cache"),
+            default_max_line_length=36,
             render_video=True
         )
 
@@ -138,6 +140,7 @@ def test_video_params(self, resolution, expected_dims, expected_font, expected_l
             output_styles_json=str(styles_path),
             output_dir=str(tmp_path / "output"),
             cache_dir=str(tmp_path / "cache"),
+            default_max_line_length=36,
             video_resolution=resolution,
             render_video=True
         )
@@ -157,6 +160,7 @@ def test_invalid_video_resolution(self, tmp_path, basic_styles_content):
             output_styles_json=str(styles_path),
             output_dir=str(tmp_path / "output"),
             cache_dir=str(tmp_path / "cache"),
+            default_max_line_length=36,
             video_resolution="invalid",
             render_video=True
         )