FEATURE: propogate improvements from CLI version to web-backend

web/backend/services/splitter.py

@@ -2,10 +2,10 @@
 
 import os
 import sys
-import time
+import shutil
 from pathlib import Path
-from types import SimpleNamespace
-from typing import Dict, List, Any
+from typing import List, Dict, Any
+import time
 
 from backend.config import settings
 from backend.services.task_manager import task_manager
@@ -13,31 +13,12 @@ from backend.services.file_manager import FileManager
 from backend.models.request import TracklistEntry
 from backend.models.response import TaskStatus
 
-# Import central defaults
-from audio_splitter.defaults import (
-    DEFAULT_FORMAT,
-    DEFAULT_OUTPUT_TEMPLATE,
-    DEFAULT_REPLACEMENT_CHAR,
-    DEFAULT_BAD_CHARS,
-    DEFAULT_ALBUM,
-    DEFAULT_COMMENT,
-    DEFAULT_NO_COMMENT,
-    DEFAULT_COMMENT_STREAM,
-    DEFAULT_MERGE_COMMENTS,
-    DEFAULT_COMMENT_SEPARATOR,
-    DEFAULT_DROP_VIDEO,
-    DEFAULT_DROP_SUBS,
-    DEFAULT_NUMBER_TRACKS,
-    DEFAULT_REPLACE_BAD_CHARS,
-    DEFAULT_SKIP_EXISTING,
-    DEFAULT_DELETE_ORIGINAL,
-    DEFAULT_TRANSCODE_TO,
-)
-
 
 def run_split_task(task_id: str, tracklist: List[TracklistEntry], options: Dict[str, Any]) -> None:
     try:
-        task_manager.update_task_with_progress(task_id, progress=5, message="Initializing...")
+        task_manager.update_task_with_progress(
+            task_id, progress=5, message="Initializing..."
+        )
 
         input_path = FileManager.get_input_path(task_id)
         if not input_path or not input_path.exists():
@@ -45,7 +26,6 @@ def run_split_task(task_id: str, tracklist: List[TracklistEntry], options: Dict[
 
         output_dir = FileManager.ensure_output_dir(task_id)
 
-        # Convert tracklist to dicts (CLI format)
         tracks = []
         for entry in tracklist:
             track_dict = {
@@ -58,25 +38,30 @@ def run_split_task(task_id: str, tracklist: List[TracklistEntry], options: Dict[
             }
             tracks.append(track_dict)
 
-        # Build args namespace using defaults where options not provided
+        from types import SimpleNamespace
+
+        # Build args namespace using the new fields
         args = SimpleNamespace(
-            format=options.get("format", DEFAULT_FORMAT),
-            transcode_to=options.get("transcode_to", DEFAULT_TRANSCODE_TO),
-            drop_video=options.get("drop_video", DEFAULT_DROP_VIDEO),
-            drop_subs=options.get("drop_subs", DEFAULT_DROP_SUBS),
-            number_tracks=options.get("number_tracks", DEFAULT_NUMBER_TRACKS),
-            replace_bad_chars=options.get("replace_bad_chars", DEFAULT_REPLACE_BAD_CHARS),
-            replacement_char=options.get("replacement_char", DEFAULT_REPLACEMENT_CHAR),
-            bad_chars=options.get("bad_chars", DEFAULT_BAD_CHARS),
-            skip_existing=options.get("skip_existing", DEFAULT_SKIP_EXISTING),
-            output_template=options.get("output_template", DEFAULT_OUTPUT_TEMPLATE),
-            album=options.get("album", DEFAULT_ALBUM),
-            comment=options.get("comment", DEFAULT_COMMENT),
-            no_comment=options.get("no_comment", DEFAULT_NO_COMMENT),
-            comment_stream=options.get("comment_stream", DEFAULT_COMMENT_STREAM),
-            merge_comments=options.get("merge_comments", DEFAULT_MERGE_COMMENTS),
-            comment_separator=options.get("comment_separator", DEFAULT_COMMENT_SEPARATOR),
-            delete_original=DEFAULT_DELETE_ORIGINAL,  # never delete in web
+            container=options.get("container"),  # None -> auto-detect
+            audio_codec=options.get("audio_codec", "copy"),
+            video_codec=options.get("video_codec", "copy"),
+            subtitle_codec=options.get("subtitle_codec", "copy"),
+            video_quality=options.get("video_quality"),
+            drop_video=options.get("drop_video", False),
+            drop_subs=options.get("drop_subs", False),
+            number_tracks=options.get("number_tracks", False),
+            replace_bad_chars=options.get("replace_bad_chars", False),
+            replacement_char=options.get("replacement_char", "_"),
+            bad_chars=options.get("bad_chars", r'!@#№$;:%^&?*(){}[]\/<>+=~`\' '),
+            skip_existing=options.get("skip_existing", False),
+            output_template=options.get("output_template", "%an-%tn.%ext"),
+            album=options.get("album", None),
+            comment=options.get("comment", None),
+            no_comment=options.get("no_comment", False),
+            comment_stream=options.get("comment_stream", None),
+            merge_comments=options.get("merge_comments", False),
+            comment_separator=options.get("comment_separator", "; "),
+            delete_original=False,
         )
 
         project_root = Path(__file__).parent.parent.parent.parent
@@ -85,7 +70,25 @@ def run_split_task(task_id: str, tracklist: List[TracklistEntry], options: Dict[
 
         from audio_splitter.core import split_audio
 
-        task_manager.update_task_with_progress(task_id, progress=10, message="Starting split...")
+        # Detect attached picture and extract if applicable
+        cover_image_path = None
+        if not args.drop_video:
+            from audio_splitter.ffmpeg import is_attached_picture, extract_cover_image, get_stream_info
+            input_path_str = str(input_path)
+            stream_info = get_stream_info(input_path_str)
+            if stream_info.get('has_video') and is_attached_picture(input_path_str):
+                cover_image_path = output_dir / 'cover.png'
+                if extract_cover_image(input_path_str, str(cover_image_path)):
+                    print(f"Extracted cover image for task {task_id}")
+                else:
+                    cover_image_path = None
+
+        # Attach cover image to args
+        args.cover_image = str(cover_image_path) if cover_image_path else None
+
+        task_manager.update_task_with_progress(
+            task_id, progress=10, message="Starting split..."
+        )
 
         split_audio(str(input_path), str(output_dir), tracks, args)