FEATURE: propogate improvements from CLI version to web-backend

web/backend/api/__init__.py

@@ -1,3 +1,3 @@
 """API route handlers."""
 
-from . import upload, split, split_file, status, download, formats, info
+from . import upload, split, split_file, status, download, websocket, formats, info

web/backend/api/split.py

@@ -1,4 +1,4 @@
-"""Split task endpoint."""
+"""Split task endpoint (JSON tracklist)."""
 
 from fastapi import APIRouter, HTTPException, BackgroundTasks
 
@@ -24,6 +24,29 @@ async def start_split(request: SplitRequest, background_tasks: BackgroundTasks):
             detail=f"Task {task_id} is already {status['status']}"
         )
 
+    # Build options dict from request
+    options = {
+        "container": request.container,
+        "audio_codec": request.audio_codec,
+        "video_codec": request.video_codec,
+        "subtitle_codec": request.subtitle_codec,
+        "video_quality": request.video_quality,
+        "drop_video": request.drop_video,
+        "drop_subs": request.drop_subs,
+        "number_tracks": request.number_tracks,
+        "replace_bad_chars": request.replace_bad_chars,
+        "replacement_char": request.replacement_char,
+        "bad_chars": request.bad_chars,
+        "skip_existing": request.skip_existing,
+        "output_template": request.output_template,
+        "album": request.album,
+        "comment": request.comment,
+        "no_comment": request.no_comment,
+        "comment_stream": request.comment_stream,
+        "merge_comments": request.merge_comments,
+        "comment_separator": request.comment_separator,
+    }
+
     task_manager.update_task(
         task_id,
         status=TaskStatus.PROCESSING,
@@ -31,9 +54,14 @@ async def start_split(request: SplitRequest, background_tasks: BackgroundTasks):
         message="Preparing to split..."
     )
 
-    background_tasks.add_task(run_split_task, task_id, request.tracklist, request.options)
+    background_tasks.add_task(
+        run_split_task,
+        task_id,
+        request.tracklist,
+        options
+    )
 
     return SplitResponse(
         task_id=task_id,
         status=TaskStatus.PROCESSING
-    )
+    )

web/backend/api/split_file.py

@@ -2,6 +2,7 @@
 
 import json
 from pathlib import Path
+from typing import Optional
 
 from fastapi import APIRouter, UploadFile, File, Form, HTTPException, BackgroundTasks
 
@@ -13,7 +14,6 @@ from backend.models.response import SplitResponse, TaskStatus
 
 router = APIRouter(prefix="/api", tags=["split"])
 
-# Import CLI tracklist parsing functions
 from audio_splitter.tracklist import parse_format, read_tracklist
 
 
@@ -22,29 +22,33 @@ async def split_from_file(
     background_tasks: BackgroundTasks,
     task_id: str = Form(...),
     tracklist_file: UploadFile = File(...),
-    options: str = Form("{}"),
+    # New fields
+    container: Optional[str] = Form(None),
+    audio_codec: str = Form("copy"),
+    video_codec: str = Form("copy"),
+    subtitle_codec: str = Form("copy"),
+    video_quality: Optional[int] = Form(None),
+    drop_video: bool = Form(False),
+    drop_subs: bool = Form(False),
+    number_tracks: bool = Form(False),
+    replace_bad_chars: bool = Form(False),
+    replacement_char: str = Form("_"),
+    bad_chars: str = Form(r'!@#№$;:%^&?*(){}[]\/<>+=~`\' '),
+    skip_existing: bool = Form(False),
+    output_template: str = Form("%an-%tn.%ext"),
+    album: Optional[str] = Form(None),
+    comment: Optional[str] = Form(None),
+    no_comment: bool = Form(False),
+    comment_stream: Optional[int] = Form(None),
+    merge_comments: bool = Form(False),
+    comment_separator: str = Form("; "),
+    options: str = Form("{}"),  # backward-compatible, but we now use explicit fields
     tracklist_format: str = Form("%ts %tn - %an"),
 ):
-    """
-    Start a splitting task using a tracklist file.
-
-    This endpoint accepts a plain text tracklist file (like the CLI does)
-    and parses it using the same logic.
-
-    Args:
-        task_id: Task ID from upload.
-        tracklist_file: Tracklist file (text/plain).
-        options: JSON string of all CLI options.
-        tracklist_format: Format string for parsing (default: "%ts %tn - %an").
-
-    Returns:
-        SplitResponse with task_id and status.
-    """
     # Validate task exists
     if not task_manager.has_task(task_id):
         raise HTTPException(status_code=404, detail=f"Task {task_id} not found")
 
-    # Check if task is already processing
     status = task_manager.get_status(task_id)
     if status and status["status"] in (TaskStatus.PROCESSING, TaskStatus.DONE):
         raise HTTPException(
@@ -52,14 +56,12 @@ async def split_from_file(
             detail=f"Task {task_id} is already {status['status']}"
         )
 
-    # Validate file type
     if not tracklist_file.filename.endswith(('.txt', '.text')):
         raise HTTPException(
             status_code=400,
             detail="Tracklist file must be a text file (.txt or .text)"
         )
 
-    # Read and parse the tracklist file
     try:
         content = await tracklist_file.read()
         text = content.decode('utf-8')
@@ -75,17 +77,13 @@ async def split_from_file(
             detail="Tracklist file is empty"
         )
 
-    # Parse the tracklist using CLI logic
+    # Parse tracklist using CLI logic
     try:
-        # Parse the format string into tokens
         tokens = parse_format(tracklist_format)
-
-        # Write the content to a temporary file (read_tracklist expects a file path)
         temp_tracklist_path = settings.temp_dir / task_id / "tracklist.txt"
         temp_tracklist_path.parent.mkdir(parents=True, exist_ok=True)
         temp_tracklist_path.write_text(text, encoding='utf-8')
 
-        # Parse the tracklist file using CLI logic
         tracklist_dicts = read_tracklist(str(temp_tracklist_path), tokens)
 
     except Exception as e:
@@ -100,7 +98,7 @@ async def split_from_file(
             detail="No valid tracks found in tracklist file"
         )
 
-    # Convert dicts to TracklistEntry objects (same as JSON endpoint)
+    # Convert dicts to TracklistEntry objects
     try:
         tracklist_entries = [TracklistEntry(**entry) for entry in tracklist_dicts]
     except Exception as e:
@@ -109,14 +107,28 @@ async def split_from_file(
             detail=f"Invalid tracklist data: {str(e)}"
         )
 
-    # Parse options JSON
+    # Build options dict from explicit fields (ignore `options` parameter)
-    try:
+    options = {
-        options_dict = json.loads(options)
+        "container": container,
-    except json.JSONDecodeError:
+        "audio_codec": audio_codec,
-        raise HTTPException(
+        "video_codec": video_codec,
-            status_code=400,
+        "subtitle_codec": subtitle_codec,
-            detail="Invalid JSON in options field"
+        "video_quality": video_quality,
-        )
+        "drop_video": drop_video,
+        "drop_subs": drop_subs,
+        "number_tracks": number_tracks,
+        "replace_bad_chars": replace_bad_chars,
+        "replacement_char": replacement_char,
+        "bad_chars": bad_chars,
+        "skip_existing": skip_existing,
+        "output_template": output_template,
+        "album": album,
+        "comment": comment,
+        "no_comment": no_comment,
+        "comment_stream": comment_stream,
+        "merge_comments": merge_comments,
+        "comment_separator": comment_separator,
+    }
 
     # Update task status
     task_manager.update_task(
@@ -126,12 +138,11 @@ async def split_from_file(
         message="Preparing to split..."
     )
 
-    # Start background task
     background_tasks.add_task(
         run_split_task,
         task_id,
-        tracklist_entries,  # now TracklistEntry objects
+        tracklist_entries,
-        options_dict
+        options
     )
 
     return SplitResponse(

web/backend/models/request.py

@@ -17,19 +17,16 @@ class TracklistEntry(BaseModel):
 
     @validator("ts")
     def validate_timestamp(cls, v: str) -> str:
-        """Basic timestamp validation (format and range)."""
         v = v.strip()
         if not v:
             raise ValueError("Timestamp cannot be empty")
 
-        # Check for range format (start-end)
         if "-" in v:
             parts = v.split("-", 1)
             start = parts[0].strip()
             end = parts[1].strip()
             if not start or not end:
                 raise ValueError("Invalid range format. Expected 'start-end'")
-            # Validate each part with the same logic
             for ts in [start, end]:
                 cls._validate_single_timestamp(ts)
         else:
@@ -39,7 +36,6 @@ class TracklistEntry(BaseModel):
 
     @staticmethod
     def _validate_single_timestamp(ts: str) -> None:
-        """Validate a single timestamp (mm:ss or HH:MM:SS)."""
         parts = ts.split(":")
         if len(parts) not in (2, 3):
             raise ValueError(f"Invalid timestamp format: {ts}. Expected mm:ss or HH:MM:SS")
@@ -55,4 +51,30 @@ class SplitRequest(BaseModel):
 
     task_id: str = Field(..., description="Task ID from upload")
     tracklist: List[TracklistEntry] = Field(..., description="List of tracks")
-    options: dict = Field(default_factory=dict, description="All CLI options")
+
+    # Container and codec options
+    container: Optional[str] = Field(None, description="Output container (auto-detect if None)")
+    audio_codec: Optional[str] = Field("copy", description="Audio codec (copy or encoder name)")
+    video_codec: Optional[str] = Field("copy", description="Video codec (copy or encoder name)")
+    subtitle_codec: Optional[str] = Field("copy", description="Subtitle codec (copy or encoder name)")
+    video_quality: Optional[int] = Field(None, description="Video quality (encoder-specific integer)")
+
+    # Stream handling
+    drop_video: bool = Field(False, description="Remove video streams")
+    drop_subs: bool = Field(False, description="Remove subtitle streams")
+
+    # Filename options
+    number_tracks: bool = Field(False, description="Prepend track numbers")
+    replace_bad_chars: bool = Field(False, description="Replace bad characters")
+    replacement_char: str = Field("_", description="Replacement character")
+    bad_chars: str = Field(r'!@#№$;:%^&?*(){}[]\/<>+=~`\' ', description="Bad characters to replace")
+    skip_existing: bool = Field(False, description="Skip existing files")
+    output_template: str = Field("%an-%tn.%ext", description="Output filename template")
+
+    # Metadata options
+    album: Optional[str] = Field(None, description="Album name")
+    comment: Optional[str] = Field(None, description="Comment")
+    no_comment: bool = Field(False, description="Ignore comment")
+    comment_stream: Optional[int] = Field(None, description="Comment stream index")
+    merge_comments: bool = Field(False, description="Merge all comments")
+    comment_separator: str = Field("; ", description="Separator for merged comments")

web/backend/services/__init__.py

@@ -1 +1 @@
-"""Business logic services."""
+"""Business logic services."""

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 List, Dict, Any
-from typing import Dict, List, 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),
+            container=options.get("container"),  # None -> auto-detect
-            transcode_to=options.get("transcode_to", DEFAULT_TRANSCODE_TO),
+            audio_codec=options.get("audio_codec", "copy"),
-            drop_video=options.get("drop_video", DEFAULT_DROP_VIDEO),
+            video_codec=options.get("video_codec", "copy"),
-            drop_subs=options.get("drop_subs", DEFAULT_DROP_SUBS),
+            subtitle_codec=options.get("subtitle_codec", "copy"),
-            number_tracks=options.get("number_tracks", DEFAULT_NUMBER_TRACKS),
+            video_quality=options.get("video_quality"),
-            replace_bad_chars=options.get("replace_bad_chars", DEFAULT_REPLACE_BAD_CHARS),
+            drop_video=options.get("drop_video", False),
-            replacement_char=options.get("replacement_char", DEFAULT_REPLACEMENT_CHAR),
+            drop_subs=options.get("drop_subs", False),
-            bad_chars=options.get("bad_chars", DEFAULT_BAD_CHARS),
+            number_tracks=options.get("number_tracks", False),
-            skip_existing=options.get("skip_existing", DEFAULT_SKIP_EXISTING),
+            replace_bad_chars=options.get("replace_bad_chars", False),
-            output_template=options.get("output_template", DEFAULT_OUTPUT_TEMPLATE),
+            replacement_char=options.get("replacement_char", "_"),
-            album=options.get("album", DEFAULT_ALBUM),
+            bad_chars=options.get("bad_chars", r'!@#№$;:%^&?*(){}[]\/<>+=~`\' '),
-            comment=options.get("comment", DEFAULT_COMMENT),
+            skip_existing=options.get("skip_existing", False),
-            no_comment=options.get("no_comment", DEFAULT_NO_COMMENT),
+            output_template=options.get("output_template", "%an-%tn.%ext"),
-            comment_stream=options.get("comment_stream", DEFAULT_COMMENT_STREAM),
+            album=options.get("album", None),
-            merge_comments=options.get("merge_comments", DEFAULT_MERGE_COMMENTS),
+            comment=options.get("comment", None),
-            comment_separator=options.get("comment_separator", DEFAULT_COMMENT_SEPARATOR),
+            no_comment=options.get("no_comment", False),
-            delete_original=DEFAULT_DELETE_ORIGINAL,  # never delete in web
+            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)