nfoguard/nfoguard.py

#!/usr/bin/env python3
"""
NFOGuard - Consolidated webhook server for TV and Movie import date management
Replaces webhook_server.py and media_date_cache.py with modular architecture
"""
import os
import json
import asyncio
import glob
import re
import logging
import logging.handlers
from pathlib import Path
from datetime import datetime, timezone, timedelta
from fastapi import FastAPI, HTTPException, BackgroundTasks, Request
from pydantic import BaseModel
from typing import Optional, Dict, Any, List, Set, Tuple
import threading
from concurrent.futures import ThreadPoolExecutor
import uvicorn

# Import our new modular components
from core.database import NFOGuardDatabase
from core.nfo_manager import NFOManager
from core.path_mapper import PathMapper
from clients.radarr_client import RadarrClient
from clients.sonarr_client import SonarrClient
from clients.external_clients import ExternalClientManager

# ---------------------------
# Configuration & Logging
# ---------------------------

def _setup_file_logging():
    """Setup file logging for NFOGuard"""
    log_dir = Path("/app/data/logs")
    log_dir.mkdir(parents=True, exist_ok=True)
    
    logger = logging.getLogger("NFOGuard")
    logger.setLevel(logging.DEBUG)
    
    file_handler = logging.handlers.RotatingFileHandler(
        log_dir / "nfoguard.log", maxBytes=50*1024*1024, backupCount=3
    )
    
    formatter = logging.Formatter(
        '[%(asctime)s] %(levelname)s: %(message)s',
        datefmt='%Y-%m-%dT%H:%M:%S+00:00'
    )
    file_handler.setFormatter(formatter)
    logger.addHandler(file_handler)
    return logger

def _log(level: str, msg: str):
    """Enhanced logging that writes to both console and file"""
    print(f"[{datetime.now(timezone.utc).isoformat(timespec='seconds')}] {level}: {msg}")
    
    try:
        file_logger = _setup_file_logging()
        getattr(file_logger, level.lower(), file_logger.info)(msg)
    except Exception as e:
        print(f"File logging error: {e}")

# Initialize logging
_setup_file_logging()

def _bool_env(name: str, default: bool) -> bool:
    v = os.environ.get(name)
    if v is None:
        return default
    return v.lower() in ("1", "true", "yes", "y", "on")

# ---------------------------
# Configuration
# ---------------------------

class NFOGuardConfig:
    def __init__(self):
        # Paths
        self.tv_paths = [Path(p.strip()) for p in os.environ.get("TV_PATHS", "/media/tv,/media/tv6").split(",") if p.strip()]
        self.movie_paths = [Path(p.strip()) for p in os.environ.get("MOVIE_PATHS", "/media/movies,/media/movies6").split(",") if p.strip()]
        
        # Core settings
        self.manage_nfo = _bool_env("MANAGE_NFO", True)
        self.fix_dir_mtimes = _bool_env("FIX_DIR_MTIMES", True)
        self.lock_metadata = _bool_env("LOCK_METADATA", True)
        self.debug = _bool_env("DEBUG", False)
        self.manager_brand = os.environ.get("MANAGER_BRAND", "NFOGuard")
        
        # Batching
        self.batch_delay = float(os.environ.get("BATCH_DELAY", "5.0"))
        self.max_concurrent = int(os.environ.get("MAX_CONCURRENT_SERIES", "3"))
        
        # Database
        self.db_path = Path(os.environ.get("DB_PATH", "/app/data/media_dates.db"))
        
        # Movie processing
        self.movie_priority = os.environ.get("MOVIE_PRIORITY", "import_then_digital").lower()
        self.movie_poll_mode = os.environ.get("MOVIE_POLL_MODE", "always").lower()
        self.movie_update_mode = os.environ.get("MOVIE_DATE_UPDATE_MODE", "backfill_only").lower()

config = NFOGuardConfig()

# ---------------------------
# Models
# ---------------------------

class SonarrWebhook(BaseModel):
    eventType: str
    series: Optional[Dict[str, Any]] = None
    episodes: Optional[list] = []
    episodeFile: Optional[Dict[str, Any]] = None
    isUpgrade: Optional[bool] = False

    class Config:
        extra = "allow"

class RadarrWebhook(BaseModel):
    eventType: str
    movie: Optional[Dict[str, Any]] = None
    movieFile: Optional[Dict[str, Any]] = None
    isUpgrade: Optional[bool] = False
    deletedFiles: Optional[list] = []
    remoteMovie: Optional[Dict[str, Any]] = None
    renamedMovieFiles: Optional[List[Dict[str, Any]]] = None

    class Config:
        extra = "allow"

class HealthResponse(BaseModel):
    status: str
    version: str
    uptime: str
    database_status: str
    radarr_database: Optional[Dict[str, Any]] = None

# ---------------------------
# Core Processing
# ---------------------------

class TVProcessor:
    """Handles TV series processing"""
    
    def __init__(self, db: NFOGuardDatabase, nfo_manager: NFOManager, path_mapper: PathMapper):
        self.db = db
        self.nfo_manager = nfo_manager
        self.path_mapper = path_mapper
        self.sonarr = SonarrClient(
            os.environ.get("SONARR_URL", ""),
            os.environ.get("SONARR_API_KEY", "")
        )
        self.external_clients = ExternalClientManager()
    
    def find_series_path(self, series_title: str, imdb_id: str, sonarr_path: str = None) -> Optional[Path]:
        """Find series directory path"""
        # Try webhook path first
        if sonarr_path:
            container_path = self.path_mapper.sonarr_path_to_container_path(sonarr_path)
            path_obj = Path(container_path)
            if path_obj.exists():
                return path_obj
        
        # Search by IMDb ID or title
        for media_path in config.tv_paths:
            if not media_path.exists():
                continue
            
            # Search by IMDb ID
            if imdb_id:
                pattern = str(media_path / f"*[imdb-{imdb_id}]*")
                matches = glob.glob(pattern)
                if matches:
                    return Path(matches[0])
            
            # Search by title
            if series_title:
                title_clean = series_title.lower().replace(" ", "").replace("-", "")
                for item in media_path.iterdir():
                    if item.is_dir() and "[imdb-" in item.name.lower():
                        item_clean = item.name.lower().replace(" ", "").replace("-", "")
                        if title_clean in item_clean:
                            return item
        
        return None
    
    def process_series(self, series_path: Path) -> None:
        """Process a TV series directory"""
        imdb_id = self.nfo_manager.parse_imdb_from_path(series_path)
        if not imdb_id:
            _log("ERROR", f"No IMDb ID found in series path: {series_path}")
            return
        
        _log("INFO", f"Processing TV series: {series_path.name}")
        
        # Update database
        self.db.upsert_series(imdb_id, str(series_path))
        
        # Find video files
        disk_episodes = self._find_disk_episodes(series_path)
        _log("INFO", f"Found {len(disk_episodes)} episodes on disk")
        
        # Get episode dates
        episode_dates = self._gather_episode_dates(series_path, imdb_id, disk_episodes)
        
        # Process episodes
        for (season, episode), (aired, dateadded, source) in episode_dates.items():
            if (season, episode) in disk_episodes:
                # Create NFO
                if config.manage_nfo:
                    self.nfo_manager.create_episode_nfo(
                        series_path / f"Season {season:02d}",
                        season, episode, aired, dateadded, source, config.lock_metadata
                    )
                
                # Update file mtimes
                if config.fix_dir_mtimes and dateadded:
                    video_files = disk_episodes[(season, episode)]
                    for video_file in video_files:
                        self.nfo_manager.set_file_mtime(video_file, dateadded)
                
                # Save to database
                self.db.upsert_episode_date(imdb_id, season, episode, aired, dateadded, source, True)
        
        # Create season/tvshow NFOs
        if config.manage_nfo:
            seasons_processed = set()
            for (season, episode) in disk_episodes.keys():
                if season not in seasons_processed:
                    season_dir = series_path / f"Season {season:02d}"
                    self.nfo_manager.create_season_nfo(season_dir, season)
                    seasons_processed.add(season)
            
            self.nfo_manager.create_tvshow_nfo(series_path, imdb_id)
        
        _log("INFO", f"Completed processing TV series: {series_path.name}")
    
    def _find_disk_episodes(self, series_path: Path) -> Dict[Tuple[int, int], List[Path]]:
        """Find all episode video files on disk"""
        disk_episodes = {}
        video_exts = (".mkv", ".mp4", ".avi", ".mov", ".m4v")
        
        for season_dir in series_path.iterdir():
            if not (season_dir.is_dir() and season_dir.name.lower().startswith("season ")):
                continue
            
            try:
                season_num = int(season_dir.name.split()[-1])
            except Exception:
                continue
            
            for video_file in season_dir.iterdir():
                if video_file.is_file() and video_file.suffix.lower() in video_exts:
                    match = re.search(r"S(\d{2})E(\d{2})", video_file.name, re.IGNORECASE)
                    if match:
                        file_season, file_episode = int(match.group(1)), int(match.group(2))
                        key = (season_num, file_episode)  # Use directory season number
                        if key not in disk_episodes:
                            disk_episodes[key] = []
                        disk_episodes[key].append(video_file)
        
        return disk_episodes
    
    def _gather_episode_dates(self, series_path: Path, imdb_id: str, disk_episodes: Dict) -> Dict:
        """Gather episode dates from various sources"""
        episode_dates = {}
        
        # Check cache first
        cached_episodes = self.db.get_series_episodes(imdb_id, has_video_file_only=True)
        for ep in cached_episodes:
            key = (ep["season"], ep["episode"])
            if key in disk_episodes:  # Only use cached data for episodes we have
                episode_dates[key] = (ep["aired"], ep["dateadded"], ep["source"])
        
        # Find missing episodes
        cached_keys = set(episode_dates.keys())
        missing_keys = set(disk_episodes.keys()) - cached_keys
        
        if not missing_keys:
            _log("INFO", "All episodes found in cache")
            return episode_dates
        
        _log("INFO", f"Querying APIs for {len(missing_keys)} missing episodes")
        
        # Query Sonarr for missing episodes
        if self.sonarr.enabled:
            series = self.sonarr.series_by_imdb(imdb_id)
            if series:
                series_id = series.get("id")
                if series_id:
                    episodes = self.sonarr.episodes_for_series(series_id)
                    for ep in episodes:
                        season_num = ep.get("seasonNumber")
                        episode_num = ep.get("episodeNumber")
                        
                        if not isinstance(season_num, int) or not isinstance(episode_num, int):
                            continue
                        
                        key = (season_num, episode_num)
                        if key not in missing_keys:
                            continue
                        
                        # Get dates
                        aired = self._parse_date_to_iso(ep.get("airDateUtc"))
                        dateadded = None
                        source = "sonarr:episode.airDateUtc"
                        
                        # Try to get import history
                        episode_id = ep.get("id")
                        if episode_id:
                            import_date = self.sonarr.get_episode_import_history(episode_id)
                            if import_date:
                                dateadded = self._parse_date_to_iso(import_date)
                                source = "sonarr:history.import"
                        
                        if not dateadded:
                            dateadded = aired
                        
                        if aired or dateadded:
                            episode_dates[key] = (aired, dateadded, source)
        
        # Fill remaining gaps with external APIs
        remaining_keys = missing_keys - set(episode_dates.keys())
        if remaining_keys and self.external_clients.tmdb.enabled:
            tmdb_movie = self.external_clients.tmdb.find_by_imdb(imdb_id)
            if tmdb_movie:
                tv_id = tmdb_movie.get("id")
                if tv_id:
                    seasons_needed = set(season for season, episode in remaining_keys)
                    for season_num in seasons_needed:
                        tmdb_episodes = self.external_clients.tmdb.get_tv_season_episodes(tv_id, season_num)
                        for ep_num, air_date in tmdb_episodes.items():
                            key = (season_num, ep_num)
                            if key in remaining_keys:
                                aired = self._parse_date_to_iso(air_date)
                                episode_dates[key] = (aired, aired, "tmdb:air_date")
        
        return episode_dates
    
    def _parse_date_to_iso(self, date_str: str) -> Optional[str]:
        """Parse date string to ISO format"""
        if not date_str:
            return None
        try:
            if len(date_str) == 10 and date_str[4] == "-":
                dt = datetime.fromisoformat(date_str).replace(tzinfo=timezone.utc)
            else:
                dt = datetime.fromisoformat(date_str.replace("Z", "+00:00")).astimezone(timezone.utc)
            return dt.isoformat(timespec="seconds")
        except Exception:
            return None


class MovieProcessor:
    """Handles movie processing"""
    
    def __init__(self, db: NFOGuardDatabase, nfo_manager: NFOManager, path_mapper: PathMapper):
        self.db = db
        self.nfo_manager = nfo_manager
        self.path_mapper = path_mapper
        self.radarr = RadarrClient(
            os.environ.get("RADARR_URL", ""),
            os.environ.get("RADARR_API_KEY", "")
        )
        self.external_clients = ExternalClientManager()
    
    def find_movie_path(self, movie_title: str, imdb_id: str, radarr_path: str = None) -> Optional[Path]:
        """Find movie directory path"""
        # Try webhook path first
        if radarr_path:
            container_path = self.path_mapper.radarr_path_to_container_path(radarr_path)
            path_obj = Path(container_path)
            if path_obj.exists():
                return path_obj
        
        # Search by IMDb ID or title
        for media_path in config.movie_paths:
            if not media_path.exists():
                continue
            
            # Search by IMDb ID
            if imdb_id:
                pattern = str(media_path / f"*[imdb-{imdb_id}]*")
                matches = glob.glob(pattern)
                if matches:
                    return Path(matches[0])
            
            # Search by title
            if movie_title:
                title_clean = movie_title.lower().replace(" ", "").replace("-", "")
                for item in media_path.iterdir():
                    if item.is_dir() and "[imdb-" in item.name.lower():
                        item_clean = item.name.lower().replace(" ", "").replace("-", "")
                        if title_clean in item_clean:
                            return item
        
        return None
    
    def process_movie(self, movie_path: Path) -> None:
        """Process a movie directory"""
        imdb_id = self.nfo_manager.parse_imdb_from_path(movie_path)
        if not imdb_id:
            _log("ERROR", f"No IMDb ID found in movie path: {movie_path}")
            return
        
        _log("INFO", f"Processing movie: {movie_path.name}")
        
        # Update database
        self.db.upsert_movie(imdb_id, str(movie_path))
        
        # Check for video files
        video_exts = (".mkv", ".mp4", ".avi", ".mov", ".m4v")
        has_video = any(f.is_file() and f.suffix.lower() in video_exts for f in movie_path.iterdir())
        
        if not has_video:
            _log("WARNING", f"No video files found in: {movie_path}")
            self.db.upsert_movie_dates(imdb_id, None, None, None, False)
            return
        
        # Get existing dates
        existing = self.db.get_movie_dates(imdb_id)
        
        # Determine if we should query APIs
        should_query = (
            config.movie_poll_mode == "always" or
            (config.movie_poll_mode == "if_missing" and not existing) or
            (config.movie_poll_mode == "if_missing" and existing and existing.get("source") == "file:mtime")
        )
        
        # Decide dates
        dateadded, source, released = self._decide_movie_dates(imdb_id, movie_path, should_query, existing)
        
        # Create NFO
        if config.manage_nfo:
            self.nfo_manager.create_movie_nfo(
                movie_path, imdb_id, dateadded, released, source, config.lock_metadata
            )
        
        # Update file mtimes
        if config.fix_dir_mtimes and dateadded and dateadded != "MANUAL_REVIEW_NEEDED":
            self.nfo_manager.update_movie_files_mtime(movie_path, dateadded)
        
        # Save to database
        self.db.upsert_movie_dates(imdb_id, released, dateadded, source, True)
        
        _log("INFO", f"Completed processing movie: {movie_path.name} (source: {source})")
    
    def _decide_movie_dates(self, imdb_id: str, movie_path: Path, should_query: bool, existing: Optional[Dict]) -> Tuple[str, str, Optional[str]]:
        """Decide movie dates based on configuration and available data"""
        if not should_query and existing:
            return existing["dateadded"], existing["source"], existing.get("released")
        
        # Query Radarr for movie info
        radarr_movie = None
        if should_query and self.radarr.api_key:
            radarr_movie = self.radarr.movie_by_imdb(imdb_id)
        
        released = None
        if radarr_movie:
            released = self._parse_date_to_iso(radarr_movie.get("inCinemas"))
        
        # Try import history first if configured
        if config.movie_priority == "import_then_digital":
            if radarr_movie:
                movie_id = radarr_movie.get("id")
                if movie_id:
                    import_date, import_source = self.radarr.get_movie_import_date(movie_id)
                    if import_date:
                        return import_date, import_source, released
            
            # Fall back to digital release
            digital_date, digital_source = self._get_digital_release_date(imdb_id)
            if digital_date:
                return digital_date, digital_source, released
        
        else:  # digital_then_import
            # Try digital release first
            digital_date, digital_source = self._get_digital_release_date(imdb_id)
            if digital_date:
                return digital_date, digital_source, released
            
            # Fall back to import history
            if radarr_movie:
                movie_id = radarr_movie.get("id")
                if movie_id:
                    import_date, import_source = self.radarr.get_movie_import_date(movie_id)
                    if import_date:
                        return import_date, import_source, released
        
        # Last resort: file mtime
        return self._get_file_mtime_date(movie_path)
    
    def _get_digital_release_date(self, imdb_id: str) -> Tuple[Optional[str], str]:
        """Get digital release date from external sources"""
        digital_release = self.external_clients.get_earliest_digital_release(imdb_id)
        if digital_release:
            return digital_release[0], digital_release[1]
        return None, "digital:none"
    
    def _get_file_mtime_date(self, movie_path: Path) -> Tuple[str, str, Optional[str]]:
        """Get date from file modification time as last resort"""
        video_exts = (".mkv", ".mp4", ".avi", ".mov", ".m4v")
        newest_mtime = None
        
        for file_path in movie_path.iterdir():
            if file_path.is_file() and file_path.suffix.lower() in video_exts:
                try:
                    mtime = file_path.stat().st_mtime
                    if newest_mtime is None or mtime > newest_mtime:
                        newest_mtime = mtime
                except Exception:
                    continue
        
        if newest_mtime:
            try:
                iso_date = datetime.fromtimestamp(newest_mtime, tz=timezone.utc).isoformat(timespec="seconds")
                return iso_date, "file:mtime", None
            except Exception:
                pass
        
        return "MANUAL_REVIEW_NEEDED", "manual_review_required", None
    
    def _parse_date_to_iso(self, date_str: str) -> Optional[str]:
        """Parse date string to ISO format"""
        if not date_str:
            return None
        try:
            if len(date_str) == 10 and date_str[4] == "-":
                dt = datetime.fromisoformat(date_str).replace(tzinfo=timezone.utc)
            else:
                dt = datetime.fromisoformat(date_str.replace("Z", "+00:00")).astimezone(timezone.utc)
            return dt.isoformat(timespec="seconds")
        except Exception:
            return None


# ---------------------------
# Batching System
# ---------------------------

class WebhookBatcher:
    """Batches webhook events to avoid processing storms"""
    
    def __init__(self):
        self.pending: Dict[str, Dict] = {}
        self.timers: Dict[str, threading.Timer] = {}
        self.processing: Set[str] = set()
        self.lock = threading.Lock()
        self.executor = ThreadPoolExecutor(max_workers=config.max_concurrent)
    
    def add_webhook(self, key: str, webhook_data: Dict, media_type: str):
        """Add webhook to batch queue"""
        with self.lock:
            if key in self.timers:
                self.timers[key].cancel()
            
            webhook_data['media_type'] = media_type
            self.pending[key] = webhook_data
            _log("INFO", f"Batched {media_type} webhook for {key}")
            
            timer = threading.Timer(config.batch_delay, self._process_item, args=[key])
            self.timers[key] = timer
            timer.start()
    
    def _process_item(self, key: str):
        """Process a batched item"""
        with self.lock:
            if key in self.processing or key not in self.pending:
                return
            self.processing.add(key)
            webhook_data = self.pending.pop(key)
            self.timers.pop(key, None)
        
        try:
            self.executor.submit(self._process_sync, key, webhook_data)
        except Exception as e:
            _log("ERROR", f"Error submitting processing for {key}: {e}")
            with self.lock:
                self.processing.discard(key)
    
    def _process_sync(self, key: str, webhook_data: Dict):
        """Synchronous processing of webhook data"""
        try:
            media_type = webhook_data.get('media_type')
            path_str = webhook_data.get('path')
            
            if not path_str:
                _log("ERROR", f"No path found for {media_type} {key}")
                return
            
            path_obj = Path(path_str)
            if not path_obj.exists():
                _log("ERROR", f"Path does not exist: {path_obj}")
                return
            
            # Process based on media type
            if media_type == 'tv':
                tv_processor.process_series(path_obj)
            elif media_type == 'movie':
                movie_processor.process_movie(path_obj)
            else:
                _log("ERROR", f"Unknown media type: {media_type}")
        
        except Exception as e:
            _log("ERROR", f"Error processing {media_type} {key}: {e}")
        finally:
            with self.lock:
                self.processing.discard(key)
    
    def get_status(self) -> Dict:
        """Get batch queue status"""
        with self.lock:
            return {
                "pending_items": list(self.pending.keys()),
                "processing_items": list(self.processing),
                "pending_count": len(self.pending),
                "processing_count": len(self.processing)
            }


# ---------------------------
# FastAPI Application
# ---------------------------

# Get version
try:
    version = (Path(__file__).parent / "VERSION").read_text().strip()
except:
    version = "0.1.0"

app = FastAPI(
    title="NFOGuard",
    description="Webhook server for preserving media import dates",
    version=version
)

start_time = datetime.now(timezone.utc)

# Initialize components
db = NFOGuardDatabase(config.db_path)
nfo_manager = NFOManager(config.manager_brand)
path_mapper = PathMapper()
tv_processor = TVProcessor(db, nfo_manager, path_mapper)
movie_processor = MovieProcessor(db, nfo_manager, path_mapper)
batcher = WebhookBatcher()

# ---------------------------
# Webhook Handlers
# ---------------------------

async def _read_payload(request: Request) -> dict:
    """Read webhook payload from request"""
    content_type = (request.headers.get("content-type") or "").lower()
    try:
        if "application/json" in content_type:
            return await request.json()
        form = await request.form()
        if "payload" in form:
            return json.loads(form["payload"])
        return dict(form)
    except Exception as e:
        _log("ERROR", f"Failed to read webhook payload: {e}")
        return {}

@app.post("/webhook/sonarr")
async def sonarr_webhook(request: Request, background_tasks: BackgroundTasks):
    """Handle Sonarr webhooks"""
    try:
        payload = await _read_payload(request)
        if not payload:
            raise HTTPException(status_code=422, detail="Empty Sonarr payload")
        
        webhook = SonarrWebhook(**payload)
        _log("INFO", f"Received Sonarr webhook: {webhook.eventType}")
        
        if webhook.eventType not in ["Download", "Upgrade", "Rename"]:
            return {"status": "ignored", "reason": f"Event type {webhook.eventType} not processed"}
        
        if not webhook.series:
            return {"status": "ignored", "reason": "No series data"}
        
        series_info = webhook.series
        series_title = series_info.get("title", "")
        imdb_id = series_info.get("imdbId", "").replace("tt", "").strip()
        if imdb_id:
            imdb_id = f"tt{imdb_id}"
        sonarr_path = series_info.get("path", "")
        
        if not imdb_id:
            _log("ERROR", f"No IMDb ID for series: {series_title}")
            return {"status": "error", "reason": "No IMDb ID"}
        
        # Find series path
        series_path = tv_processor.find_series_path(series_title, imdb_id, sonarr_path)
        if not series_path:
            _log("ERROR", f"Could not find series directory: {series_title} ({imdb_id})")
            return {"status": "error", "reason": "Series directory not found"}
        
        # Add to batch queue
        webhook_dict = {
            'path': str(series_path),
            'series_info': series_info,
            'event_type': webhook.eventType
        }
        batcher.add_webhook(imdb_id, webhook_dict, 'tv')
        
        return {"status": "accepted", "message": "Sonarr webhook queued"}
        
    except Exception as e:
        _log("ERROR", f"Sonarr webhook error: {e}")
        raise HTTPException(status_code=422, detail=f"Invalid webhook: {e}")

@app.post("/webhook/radarr") 
async def radarr_webhook(request: Request, background_tasks: BackgroundTasks):
    """Handle Radarr webhooks"""
    try:
        payload = await _read_payload(request)
        if not payload:
            raise HTTPException(status_code=422, detail="Empty Radarr payload")
        
        webhook = RadarrWebhook(**payload)
        _log("INFO", f"Received Radarr webhook: {webhook.eventType}")
        
        if webhook.eventType not in ["Download", "Upgrade", "Rename", "Test"]:
            return {"status": "ignored", "reason": f"Event type {webhook.eventType} not processed"}
        
        if webhook.eventType == "Test":
            return {"status": "ok", "message": "Test received"}
        
        if not webhook.movie:
            return {"status": "ignored", "reason": "No movie data"}
        
        movie_info = webhook.movie
        movie_title = movie_info.get("title", "")
        imdb_id = (movie_info.get("imdbId") or "").strip()
        if imdb_id:
            imdb_id = f"tt{imdb_id.replace('tt','')}"
        radarr_path = movie_info.get("folderPath") or movie_info.get("path", "")
        
        # Find movie path
        movie_path = movie_processor.find_movie_path(movie_title, imdb_id, radarr_path)
        if not movie_path:
            _log("ERROR", f"Could not find movie directory: {movie_title} ({imdb_id})")
            return {"status": "error", "reason": "Movie directory not found"}
        
        # Extract IMDb ID from path if not in webhook
        if not imdb_id:
            imdb_id = nfo_manager.parse_imdb_from_path(movie_path)
        
        if not imdb_id:
            _log("ERROR", f"No IMDb ID available for movie: {movie_title}")
            return {"status": "error", "reason": "No IMDb ID"}
        
        # Add to batch queue
        webhook_dict = {
            'path': str(movie_path),
            'movie_info': movie_info,
            'event_type': webhook.eventType
        }
        batcher.add_webhook(imdb_id, webhook_dict, 'movie')
        
        return {"status": "accepted", "message": "Radarr webhook queued"}
        
    except Exception as e:
        _log("ERROR", f"Radarr webhook error: {e}")
        raise HTTPException(status_code=422, detail=f"Invalid webhook: {e}")

# ---------------------------
# API Endpoints  
# ---------------------------

@app.get("/health")
async def health() -> HealthResponse:
    """Health check endpoint with Radarr database status"""
    uptime = datetime.now(timezone.utc) - start_time
    
    # Check NFOGuard database
    try:
        with db.get_connection() as conn:
            conn.execute("SELECT 1").fetchone()
        db_status = "healthy"
    except Exception as e:
        db_status = f"error: {e}"
    
    # Check Radarr database if available
    radarr_db_health = None
    overall_status = "healthy" if db_status == "healthy" else "degraded"
    
    # Get Radarr client with database access
    radarr_client = client_manager.get_radarr_client()
    if radarr_client and hasattr(radarr_client, 'db_client') and radarr_client.db_client:
        try:
            radarr_db_health = radarr_client.db_client.health_check()
            if radarr_db_health["status"] != "healthy":
                overall_status = "degraded"
        except Exception as e:
            radarr_db_health = {
                "status": "error",
                "error": str(e),
                "tested_at": datetime.now(timezone.utc).isoformat(timespec="seconds")
            }
            overall_status = "degraded"
    
    return HealthResponse(
        status=overall_status,
        version=version,
        uptime=str(uptime),
        database_status=db_status,
        radarr_database=radarr_db_health
    )

@app.get("/stats")
async def get_stats():
    """Get database statistics"""
    try:
        return db.get_stats()
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))

@app.get("/batch/status") 
async def batch_status():
    """Get batch queue status"""
    return batcher.get_status()

@app.get("/debug/movie/{imdb_id}")
async def debug_movie_import_date(imdb_id: str):
    """Debug endpoint to analyze movie import date detection"""
    try:
        if not imdb_id.startswith("tt"):
            imdb_id = f"tt{imdb_id}"
            
        _log("INFO", f"=== DEBUG MOVIE IMPORT DATE: {imdb_id} ===")
        
        if not (os.environ.get("RADARR_URL") and os.environ.get("RADARR_API_KEY")):
            return {
                "error": "Radarr not configured", 
                "imdb_id": imdb_id,
                "radarr_configured": False
            }
        
        # Create Radarr client
        from clients.radarr_client import RadarrClient
        radarr_client = RadarrClient(
            os.environ.get("RADARR_URL"),
            os.environ.get("RADARR_API_KEY")
        )
        
        # Look up movie
        movie_obj = radarr_client.movie_by_imdb(imdb_id)
        if not movie_obj:
            return {
                "error": f"Movie not found in Radarr for IMDb ID {imdb_id}",
                "imdb_id": imdb_id,
                "radarr_configured": True,
                "movie_found": False
            }
            
        movie_id = movie_obj.get("id")
        movie_title = movie_obj.get("title")
        
        _log("INFO", f"Found movie: {movie_title} (Radarr ID: {movie_id})")
        
        # Get import date with detailed logging - use the optimized method
        import_date = radarr_client.earliest_import_event_optimized(movie_id)
        if import_date:
            source = "radarr:history.import.optimized"
        else:
            # Fallback to file dateAdded
            import_date = radarr_client.earliest_file_dateadded(movie_id)
            source = "radarr:file.dateAdded" if import_date else "no_date_found"
        
        # Get movie files
        movie_files = radarr_client.movie_files(movie_id)
        file_info = []
        for f in movie_files:
            file_info.append({
                "id": f.get("id"),
                "relativePath": f.get("relativePath"),
                "dateAdded": f.get("dateAdded"),
                "quality": f.get("quality", {}).get("quality", {}).get("name", "Unknown")
            })
        
        return {
            "imdb_id": imdb_id,
            "radarr_configured": True,
            "movie_found": True,
            "movie_title": movie_title,
            "movie_id": movie_id,
            "detected_import_date": import_date,
            "import_source": source,
            "movie_files": file_info,
            "debug_info": {
                "radarr_url": os.environ.get("RADARR_URL"),
                "movie_digital_release": movie_obj.get("digitalRelease"),
                "movie_in_cinemas": movie_obj.get("inCinemas"),
                "movie_physical_release": movie_obj.get("physicalRelease"),
                "movie_folder_path": movie_obj.get("folderPath")
            }
        }
        
    except Exception as e:
        _log("ERROR", f"Debug endpoint error for {imdb_id}: {e}")
        return {
            "error": str(e),
            "imdb_id": imdb_id,
            "success": False
        }

@app.get("/debug/movie/{imdb_id}/history")
async def debug_movie_history(imdb_id: str):
    """Detailed history analysis for a movie"""
    try:
        if not imdb_id.startswith("tt"):
            imdb_id = f"tt{imdb_id}"
            
        _log("INFO", f"=== DETAILED HISTORY ANALYSIS: {imdb_id} ===")
        
        if not (os.environ.get("RADARR_URL") and os.environ.get("RADARR_API_KEY")):
            return {"error": "Radarr not configured"}
        
        from clients.radarr_client import RadarrClient
        radarr_client = RadarrClient(
            os.environ.get("RADARR_URL"),
            os.environ.get("RADARR_API_KEY")
        )
        
        # Look up movie
        movie_obj = radarr_client.movie_by_imdb(imdb_id)
        if not movie_obj:
            return {"error": f"Movie not found for {imdb_id}"}
            
        movie_id = movie_obj.get("id")
        movie_title = movie_obj.get("title")
        
        # Get ALL history pages
        all_history = []
        page = 1
        
        while page <= 10:  # Safety limit
            data = radarr_client._get("/api/v3/history", {
                "movieId": movie_id,
                "page": page,
                "pageSize": 50,
                "sortKey": "date",
                "sortDirection": "ascending"
            })
            
            if not data:
                break
                
            records = data if isinstance(data, list) else data.get("records", [])
            if not records:
                break
                
            all_history.extend(records)
            
            if len(records) < 50:
                break
                
            page += 1
        
        # Analyze each event
        analyzed_events = []
        for event in all_history:
            event_type = event.get("eventType", "")
            date_str = event.get("date", "")
            event_data = event.get("data", {})
            
            # Get source path info
            source_path = (
                event_data.get('droppedPath', '') or
                event_data.get('sourcePath', '') or 
                event_data.get('path', '') or 
                event_data.get('sourceTitle', '')
            )
            
            # Analyze if this is a real import
            is_real, reason, date_iso = radarr_client._analyze_event_for_import(event)
            
            analyzed_events.append({
                "event_type": event_type,
                "date": date_str,
                "source_path": source_path,
                "is_real_import": is_real,
                "analysis_reason": reason,
                "parsed_date": date_iso,
                "full_data": event_data
            })
        
        # Find what our algorithm would pick
        picked_date = radarr_client.earliest_import_event_optimized(movie_id)
        
        return {
            "imdb_id": imdb_id,
            "movie_title": movie_title,
            "movie_id": movie_id,
            "total_history_events": len(all_history),
            "our_algorithm_picked": picked_date,
            "all_events": analyzed_events,
            "expected_july_date": "2025-07-07" in (picked_date or "")
        }
        
    except Exception as e:
        _log("ERROR", f"History debug error for {imdb_id}: {e}")
        return {"error": str(e)}

@app.post("/manual/scan")
async def manual_scan(background_tasks: BackgroundTasks, path: Optional[str] = None, scan_type: str = "both"):
    """Manual scan endpoint"""
    if scan_type not in ["both", "tv", "movies"]:
        raise HTTPException(status_code=400, detail="scan_type must be 'both', 'tv', or 'movies'")
    
    async def run_scan():
        paths_to_scan = []
        if path:
            paths_to_scan = [Path(path)]
        else:
            if scan_type in ["both", "tv"]:
                paths_to_scan.extend(config.tv_paths)
            if scan_type in ["both", "movies"]:  
                paths_to_scan.extend(config.movie_paths)
        
        for scan_path in paths_to_scan:
            if not scan_path.exists():
                continue
                
            if scan_type in ["both", "tv"] and scan_path in config.tv_paths:
                for item in scan_path.iterdir():
                    if item.is_dir() and nfo_manager.parse_imdb_from_path(item):
                        try:
                            tv_processor.process_series(item)
                        except Exception as e:
                            _log("ERROR", f"Failed processing TV series {item}: {e}")
            
            if scan_type in ["both", "movies"] and scan_path in config.movie_paths:
                _log("INFO", f"Scanning movies in: {scan_path}")
                movie_count = 0
                for item in scan_path.iterdir():
                    if item.is_dir() and nfo_manager.parse_imdb_from_path(item):
                        movie_count += 1
                        _log("INFO", f"Processing movie: {item.name}")
                        try:
                            movie_processor.process_movie(item)  
                        except Exception as e:
                            _log("ERROR", f"Failed processing movie {item}: {e}")
                _log("INFO", f"Completed movie scan: {movie_count} movies processed in {scan_path}")
    
    background_tasks.add_task(run_scan)
    return {"status": "started", "message": f"Manual {scan_type} scan started"}

# ---------------------------
# Main
# ---------------------------

if __name__ == "__main__":
    _log("INFO", "Starting NFOGuard")
    _log("INFO", f"Version: {version}")
    _log("INFO", f"TV paths: {[str(p) for p in config.tv_paths]}")
    _log("INFO", f"Movie paths: {[str(p) for p in config.movie_paths]}")
    _log("INFO", f"Database: {config.db_path}")
    _log("INFO", f"Config: manage_nfo={config.manage_nfo}, fix_mtimes={config.fix_dir_mtimes}")
    _log("INFO", f"Movie priority: {config.movie_priority}")
    
    uvicorn.run(
        app,
        host="0.0.0.0", 
        port=int(os.environ.get("PORT", "8080")),
        reload=False
    )