Audiolib/backend/app/services/scanner.py

import os
import uuid
import asyncio
import logging
from datetime import datetime
from pathlib import Path
from sqlalchemy.ext.asyncio import AsyncSession
from sqlalchemy import select
from ..database import AsyncSessionLocal
from ..models.library import Library
from ..models.media_item import LibraryItem, BookFile, Chapter
from ..models.session import ScanJob

logger = logging.getLogger(__name__)

AUDIO_EXTENSIONS = {".mp3", ".wav", ".m4a", ".flac", ".ogg", ".aac", ".m4b", ".opus"}


def _get_audio_duration(file_path: str) -> float:
    try:
        from mutagen import File as MutagenFile
        audio = MutagenFile(file_path)
        if audio and audio.info:
            return float(audio.info.length)
    except Exception:
        pass
    return 0.0


def _get_file_size(file_path: str) -> int:
    try:
        return os.path.getsize(file_path)
    except Exception:
        return 0


def _guess_title_from_path(folder_path: str) -> str:
    """Leitet Titel aus dem Ordnernamen ab."""
    return os.path.basename(folder_path)


def _discover_audiobook_folders(base_path: str) -> list[dict]:
    """
    Findet alle Unterordner mit Audio-Dateien.
    Jeder Ordner = ein Hörbuch (ABS-Prinzip).
    """
    books = []
    base = Path(base_path)

    if not base.exists():
        logger.warning(f"Pfad nicht gefunden: {base_path}")
        return books

    # Direkte Audio-Dateien im Root → ein "Root"-Buch
    root_audio = [f for f in base.iterdir() if f.is_file() and f.suffix.lower() in AUDIO_EXTENSIONS]
    if root_audio:
        books.append({
            "path": str(base),
            "files": [str(f) for f in sorted(root_audio)],
        })

    # Unterordner durchsuchen
    for entry in base.iterdir():
        if not entry.is_dir():
            continue
        audio_files = []
        _collect_audio_files(entry, audio_files)
        if audio_files:
            books.append({
                "path": str(entry),
                "files": sorted(audio_files),
            })

    return books


def _collect_audio_files(folder: Path, result: list):
    """Rekursiv alle Audio-Dateien sammeln."""
    try:
        for entry in sorted(folder.iterdir()):
            if entry.is_file() and entry.suffix.lower() in AUDIO_EXTENSIONS:
                result.append(str(entry))
            elif entry.is_dir():
                _collect_audio_files(entry, result)
    except PermissionError:
        pass


async def scan_library_task(library_id: str, job_id: str):
    """Hintergrund-Task: Scannt eine Library und befüllt die DB."""
    async with AsyncSessionLocal() as db:
        try:
            # Job auf "running" setzen
            job_result = await db.execute(select(ScanJob).where(ScanJob.id == job_id))
            job = job_result.scalar_one_or_none()
            if job:
                job.status = "running"
                job.started_at = datetime.utcnow()
                await db.commit()

            lib_result = await db.execute(select(Library).where(Library.id == library_id))
            lib = lib_result.scalar_one_or_none()
            if not lib:
                return

            folders = lib.folders or []
            all_books = []
            for folder_info in folders:
                folder_path = folder_info.get("fullPath", folder_info.get("full_path", ""))
                if folder_path:
                    all_books.extend(_discover_audiobook_folders(folder_path))

            items_found = 0
            new_item_ids: list[str] = []
            for book_info in all_books:
                folder_path = book_info["path"]
                audio_files = book_info["files"]

                # Existiert schon?
                existing = await db.execute(
                    select(LibraryItem).where(
                        LibraryItem.library_id == library_id,
                        LibraryItem.path == folder_path,
                    )
                )
                existing_item = existing.scalar_one_or_none()

                total_duration = sum(_get_audio_duration(f) for f in audio_files)
                total_size = sum(_get_file_size(f) for f in audio_files)

                if existing_item:
                    existing_item.duration_seconds = total_duration
                    existing_item.size_bytes = total_size
                    existing_item.num_files = len(audio_files)
                    existing_item.is_missing = False
                    existing_item.updated_at = datetime.utcnow()
                    item = existing_item
                else:
                    item_id = str(uuid.uuid4())
                    title = _guess_title_from_path(folder_path)
                    item = LibraryItem(
                        id=item_id,
                        library_id=library_id,
                        media_type=lib.media_type,
                        path=folder_path,
                        ino=str(os.stat(folder_path).st_ino) if os.path.exists(folder_path) else "",
                        title=title,
                        duration_seconds=total_duration,
                        size_bytes=total_size,
                        num_files=len(audio_files),
                        tags=["zu_prüfen"],
                    )
                    db.add(item)
                    await db.flush()

                    # BookFiles anlegen
                    for idx, file_path in enumerate(audio_files):
                        bf = BookFile(
                            library_item_id=item.id,
                            filename=os.path.basename(file_path),
                            path=file_path,
                            format=Path(file_path).suffix.lstrip(".").lower(),
                            size_bytes=_get_file_size(file_path),
                            duration_seconds=_get_audio_duration(file_path),
                            track_index=idx,
                        )
                        db.add(bf)

                    new_item_ids.append(item.id)

                items_found += 1

            await db.commit()

            # Fehlende Items markieren
            all_items_result = await db.execute(
                select(LibraryItem).where(LibraryItem.library_id == library_id)
            )
            all_items = all_items_result.scalars().all()
            found_paths = {b["path"] for b in all_books}
            for item in all_items:
                item.is_missing = item.path not in found_paths
            await db.commit()

            if job:
                job.status = "done"
                job.items_found = items_found
                job.finished_at = datetime.utcnow()
                job.progress = 1.0
                await db.commit()

            logger.info(f"Scan abgeschlossen: {items_found} Items in Library {library_id}")

            # Auto-Matching für neue Items starten
            if new_item_ids:
                from .matcher import match_audiobook
                for iid in new_item_ids:
                    asyncio.create_task(match_audiobook(iid))

        except Exception as e:
            logger.error(f"Scan-Fehler für Library {library_id}: {e}", exc_info=True)
            async with AsyncSessionLocal() as err_db:
                job_result = await err_db.execute(select(ScanJob).where(ScanJob.id == job_id))
                job = job_result.scalar_one_or_none()
                if job:
                    job.status = "error"
                    job.log = str(e)
                    job.finished_at = datetime.utcnow()
                    await err_db.commit()