Updated Episode Matcher to take hints about episodes on disc and delete flags

.gitignore

@@ -0,0 +1,4 @@
+*.pyc
+*.mkv
+
+src/.tvdb_cache/*

episode_matcher.py

@@ -7,13 +7,15 @@ This tool takes a folder of MKV files and:
 2. Moves extras to an "extras" subfolder  
 3. Renames episodes using Jellyfin naming convention
 4. Uses TVDB data to validate episode matching
+5. Detects and handles duplicate episodes
 
 Usage:
-    python episode_matcher.py <folder_path> <show_name> <season_number> [--api-key <tvdb_api_key>]
+    python episode_matcher.py <folder_path> <show_name> <season_number> [options]
 
 Examples:
     python episode_matcher.py "/path/to/episodes" "Game of Thrones" 1
-    python episode_matcher.py "/path/to/episodes" "Breaking Bad" 2 --api-key your_tvdb_api_key
+    python episode_matcher.py "/path/to/episodes" "Game of Thrones" 2 --disc-mapping "1:1-3,2:4-6,3:7-9,4:10"
+    python episode_matcher.py "/path/to/episodes" "Breaking Bad" 2 --auto-delete-duplicates --dry-run
 """
 
 import argparse
@@ -30,6 +32,41 @@ from Matcher import FileClassifier, EpisodeRenamer
 from config import config_manager
 
 
+def parse_disc_mapping(mapping_str):
+    """Parse disc mapping string like '1:1-3,2:4-6,3:7-9,4:10' into a dict."""
+    mapping = {}
+    
+    if not mapping_str:
+        return mapping
+    
+    try:
+        # Split by comma to get each disc mapping
+        disc_parts = mapping_str.split(',')
+        
+        for part in disc_parts:
+            # Split by colon to get disc:episodes
+            disc_str, episodes_str = part.strip().split(':')
+            disc_num = int(disc_str)
+            
+            # Parse episode range
+            if '-' in episodes_str:
+                # Range like "1-3"
+                start, end = episodes_str.split('-')
+                episodes = list(range(int(start), int(end) + 1))
+            else:
+                # Single episode like "10"
+                episodes = [int(episodes_str)]
+            
+            mapping[disc_num] = episodes
+        
+        return mapping
+    
+    except Exception as e:
+        print(f"Error parsing disc mapping '{mapping_str}': {e}")
+        print("Expected format: '1:1-3,2:4-6,3:7-9,4:10'")
+        sys.exit(1)
+
+
 def main():
     """Main function to run the episode matcher."""
     parser = argparse.ArgumentParser(
@@ -71,6 +108,17 @@ def main():
         help="Show detailed analysis information"
     )
     
+    parser.add_argument(
+        "--disc-mapping",
+        help="Disc to episode mapping (e.g., '1:1-3,2:4-6,3:7-9,4:10')"
+    )
+    
+    parser.add_argument(
+        "--auto-delete-duplicates",
+        action="store_true",
+        help="Automatically delete duplicate files instead of moving them to 'delete me' folder"
+    )
+    
     args = parser.parse_args()
     
     # Validate inputs
@@ -164,7 +212,17 @@ def main():
                 print(f"  - {ex['name']} ({ex['size_gb']:.2f} GB, {ex['duration_minutes']:.1f} min)")
     
     # Initialize renamer
-    renamer = EpisodeRenamer(str(folder_path), args.show_name, args.season_number)
+    disc_mapping = None
+    if args.disc_mapping:
+        print(f"Using disc mapping: {args.disc_mapping}")
+        disc_mapping = parse_disc_mapping(args.disc_mapping)
+        if args.verbose:
+            print("Parsed disc mapping:")
+            for disc, episodes in disc_mapping.items():
+                print(f"  Disc {disc}: Episodes {episodes}")
+    
+    renamer = EpisodeRenamer(str(folder_path), args.show_name, args.season_number, 
+                            disc_mapping=disc_mapping, auto_delete_duplicates=args.auto_delete_duplicates)
     
     if args.dry_run:
         print(f"\n=== DRY RUN - No files will be modified ===")
@@ -175,6 +233,20 @@ def main():
                 print(f"  - {extra['name']}")
         
         print(f"\nWould rename {len(episodes)} episode files:")
+        if args.verbose:
+            print("Episodes to be renamed:")
+            for i, ep in enumerate(episodes):
+                print(f"  {i+1}: {ep} (type: {type(ep)})")
+        
+        # Handle the case where episodes is incorrectly formatted
+        if episodes and not isinstance(episodes[0], dict):
+            print("Warning: Episodes data is in wrong format, attempting to reconstruct...")
+            # This is a temporary workaround - we need to get the actual episode data
+            # from the classifier again
+            all_files = classifier.file_info
+            episode_files = [f for f in all_files if f not in extras]
+            episodes = episode_files[:len(tvdb_episodes)] if tvdb_episodes else episode_files
+            
         # Simulate the matching process
         if tvdb_episodes:
             matched_episodes = renamer._match_episodes_by_duration(episodes, tvdb_episodes)

src/Matcher/episode_renamer.py

@@ -9,12 +9,16 @@ import re
 class EpisodeRenamer:
     """Renames episode files to Jellyfin format and moves extras to subfolder."""
     
-    def __init__(self, folder_path: str, show_name: str, season_number: int):
+    def __init__(self, folder_path: str, show_name: str, season_number: int, 
-        """Initialize with folder path, show name, and season number."""
+                 disc_mapping: Optional[Dict[int, List[int]]] = None, auto_delete_duplicates: bool = False):
+        """Initialize with folder path, show name, season number, optional disc mapping, and auto-delete option."""
         self.folder_path = Path(folder_path)
         self.show_name = show_name
         self.season_number = season_number
+        self.disc_mapping = disc_mapping  # {disc_num: [episode_numbers]}
+        self.auto_delete_duplicates = auto_delete_duplicates
         self.extras_folder = self.folder_path / "extras"
+        self.delete_folder = self.folder_path / "delete me"
     
     def _create_extras_folder(self) -> bool:
         """Create extras folder if it doesn't exist."""
@@ -25,6 +29,15 @@ class EpisodeRenamer:
             print(f"Error creating extras folder: {e}")
             return False
     
+    def _create_delete_folder(self) -> bool:
+        """Create delete me folder if it doesn't exist."""
+        try:
+            self.delete_folder.mkdir(exist_ok=True)
+            return True
+        except Exception as e:
+            print(f"Error creating delete me folder: {e}")
+            return False
+    
     def _sanitize_filename(self, filename: str) -> str:
         """Sanitize filename to remove invalid characters."""
         # Remove invalid characters for file systems
@@ -72,16 +85,110 @@ class EpisodeRenamer:
         
         return moved_files
     
+    def detect_and_move_duplicates(self, episodes: List[Dict]) -> List[str]:
+        """Detect duplicate episodes and either delete them or move them to delete me folder."""
+        processed_files = []
+        duplicates_found = []
+        
+        # Group episodes by duration (within 1 minute tolerance)
+        duration_groups = {}
+        for episode in episodes:
+            duration = episode['duration_minutes']
+            
+            # Find existing group within 1 minute tolerance
+            matching_group = None
+            for group_duration in duration_groups.keys():
+                if abs(duration - group_duration) <= 0.5:  # Tighter tolerance for exact duplicates
+                    matching_group = group_duration
+                    break
+            
+            if matching_group:
+                duration_groups[matching_group].append(episode)
+            else:
+                duration_groups[duration] = [episode]
+        
+        # Identify duplicates (groups with more than one episode)
+        for duration, group_episodes in duration_groups.items():
+            if len(group_episodes) > 1:
+                print(f"\nFound {len(group_episodes)} potential duplicates with ~{duration:.1f}min duration:")
+                
+                # Sort by disc number (keep lower disc numbers) then file size (largest first) for consistent ordering
+                group_episodes.sort(key=lambda x: (
+                    x.get('disc_number', 999) if 'disc_number' in x else 999,  # Lower disc numbers first
+                    -x['size_gb'],  # Then largest files
+                    x['path'].name  # Finally by filename for consistency
+                ))
+                
+                # Keep the first file (from earliest disc, largest size), mark others as duplicates
+                keeper = group_episodes[0]
+                duplicates = group_episodes[1:]
+                
+                disc_info = ""
+                if hasattr(keeper['path'], 'name'):
+                    # Try to extract disc info from filename for display
+                    disc_match = re.search(r'Disc\s*(\d+)', keeper['path'].name, re.IGNORECASE)
+                    if disc_match:
+                        disc_info = f" (Disc {disc_match.group(1)})"
+                
+                print(f"  Keeping: {keeper['path'].name}{disc_info} ({keeper['size_gb']:.2f}GB)")
+                
+                for duplicate in duplicates:
+                    dup_disc_info = ""
+                    if hasattr(duplicate['path'], 'name'):
+                        disc_match = re.search(r'Disc\s*(\d+)', duplicate['path'].name, re.IGNORECASE)
+                        if disc_match:
+                            dup_disc_info = f" (Disc {disc_match.group(1)})"
+                    
+                    print(f"  Duplicate: {duplicate['path'].name}{dup_disc_info} ({duplicate['size_gb']:.2f}GB)")
+                    duplicates_found.append(duplicate)
+        
+        # Handle duplicates based on auto_delete_duplicates flag
+        if self.auto_delete_duplicates:
+            # Delete duplicates directly
+            for duplicate in duplicates_found:
+                source_path = duplicate['path']
+                
+                try:
+                    source_path.unlink()  # Delete the file
+                    processed_files.append(str(source_path))
+                    print(f"Deleted duplicate: {source_path.name}")
+                    
+                except Exception as e:
+                    print(f"Error deleting {source_path.name}: {e}")
+        else:
+            # Move duplicates to delete me folder (original behavior)
+            if not self._create_delete_folder():
+                return []
+            
+            for duplicate in duplicates_found:
+                source_path = duplicate['path']
+                dest_path = self.delete_folder / source_path.name
+                
+                try:
+                    # Check if destination already exists
+                    if dest_path.exists():
+                        print(f"Warning: {dest_path.name} already exists in delete me folder, skipping.")
+                        continue
+                    
+                    shutil.move(str(source_path), str(dest_path))
+                    processed_files.append(str(dest_path))
+                    print(f"Moved to delete me: {source_path.name}")
+                    
+                except Exception as e:
+                    print(f"Error moving {source_path.name} to delete me: {e}")
+        
+        return processed_files
+    
     def _extract_disc_info(self, filename: str) -> Dict:
         """Extract disc number and track number from filename."""
-        # Look for patterns like "Disc 1", "Disc1", "D1", etc.
+        # Look for patterns like "Disc 1", "Disk 1", "Disc1", "disc-1", "DISC 1_t03"
-        disc_match = re.search(r'[Dd]isc\s*(\d+)', filename, re.IGNORECASE)
+        disc_match = re.search(r'(?i)\bdis[ck][\s._-]*(\d{1,3})(?!\d)', filename)
         disc_number = int(disc_match.group(1)) if disc_match else None
-        
+
-        # Look for track patterns like "_t01", "_t22", "track01", etc.
+        # Look for track patterns like "_t01", "_t22", "track01", allow small separators
-        track_match = re.search(r'(?:_t|track)(\d+)', filename, re.IGNORECASE)
+        track_match = re.search(r'(?i)(?:_t|track)[\s._-]?(\d+)', filename)
         track_number = int(track_match.group(1)) if track_match else None
-        
+
         return {
             'disc_number': disc_number,
             'track_number': track_number,
@@ -89,7 +196,7 @@ class EpisodeRenamer:
         }
     
     def _estimate_episodes_per_disc(self, episodes_info: List[Dict], total_episodes: int) -> Dict[int, List[int]]:
-        """Estimate which episodes are on which disc based on file distribution."""
+        """Estimate which episodes are on which disc based on file distribution and storage constraints."""
         # Group files by disc and sort within each disc
         discs = {}
         for info in episodes_info:
@@ -110,19 +217,87 @@ class EpisodeRenamer:
         sorted_disc_nums = sorted(discs.keys())
         episodes_per_disc = {}
         
-        # Assign episodes sequentially across discs
+        # Calculate storage-aware episode distribution
         episode_number = 1
         for disc_num in sorted_disc_nums:
             files_on_disc = discs[disc_num]
             episodes_on_this_disc = []
             
-            for _ in files_on_disc:
+            # Calculate total storage used by episodes on this disc
+            total_disc_size_gb = sum(info['file_info']['size_gb'] for info in files_on_disc)
+            
+            # Estimate Blu-ray capacity constraints
+            # Standard Blu-ray: ~45GB usable (48GB total - overhead)
+            # Account for extras that might also be on the disc
+            estimated_episode_capacity_gb = 45.0  # Conservative estimate
+            
+            # If this disc is near capacity, reduce episode count to be conservative
+            if total_disc_size_gb > 40.0:  # Stricter threshold: 40GB
+                # This disc is likely at capacity limits
+                # Reduce episode count by 1 if we're close to limits
+                available_episodes = max(1, len(files_on_disc) - 1)
+                available_episodes = min(available_episodes, total_episodes - episode_number + 1)
+                print(f"  Warning: Disc {disc_num} near capacity ({total_disc_size_gb:.1f}GB), reducing episode allocation")
+            elif total_disc_size_gb > 35.0:  # Warning threshold: 35GB  
+                # Show warning but don't reduce count yet
+                available_episodes = min(len(files_on_disc), total_episodes - episode_number + 1)
+                print(f"  Warning: Disc {disc_num} approaching capacity ({total_disc_size_gb:.1f}GB), may have storage constraints")
+            else:
+                available_episodes = min(len(files_on_disc), total_episodes - episode_number + 1)
+            
+            for _ in range(available_episodes):
                 if episode_number <= total_episodes:
                     episodes_on_this_disc.append(episode_number)
                     episode_number += 1
             
             episodes_per_disc[disc_num] = episodes_on_this_disc
         
+        # Diagnostics: print estimated mapping per disc with storage info
+        print("\nEstimated episodes-per-disc mapping (considering storage constraints):")
+        total_assigned = 0
+        for d in sorted(episodes_per_disc.keys()):
+            files_on_disc = [f for f in episodes_info if f['disc_number'] == d]
+            total_size = sum(f['file_info']['size_gb'] for f in files_on_disc)
+            print(f"  Disc {d}: Episodes {episodes_per_disc[d]} ({len(files_on_disc)} files, {total_size:.1f}GB)")
+            total_assigned += len(episodes_per_disc[d])
+            
+            # Flag potential storage issues
+            if len(files_on_disc) > len(episodes_per_disc[d]):
+                extra_files = len(files_on_disc) - len(episodes_per_disc[d])
+                print(f"    ⚠ Warning: {extra_files} extra files detected - may indicate storage constraints")
+        
+        # Ensure we haven't lost any episodes due to capacity constraints
+        if total_assigned < total_episodes:
+            missing_episodes = total_episodes - total_assigned
+            print(f"    ⚠ Warning: {missing_episodes} episodes unassigned due to capacity constraints")
+            
+            # Try to assign remaining episodes to discs with available space
+            remaining_episodes = []
+            for ep in range(1, total_episodes + 1):
+                assigned = False
+                for disc_episodes in episodes_per_disc.values():
+                    if ep in disc_episodes:
+                        assigned = True
+                        break
+                if not assigned:
+                    remaining_episodes.append(ep)
+            
+            # Find discs with the most available space to assign remaining episodes
+            if remaining_episodes:
+                disc_space = {}
+                for d in sorted(episodes_per_disc.keys()):
+                    files_on_disc = [f for f in episodes_info if f['disc_number'] == d]
+                    total_size = sum(f['file_info']['size_gb'] for f in files_on_disc)
+                    available_space = 45.0 - total_size  # Remaining Blu-ray capacity
+                    disc_space[d] = available_space
+                
+                # Assign remaining episodes to discs with most space
+                for ep in remaining_episodes:
+                    best_disc = max(disc_space.keys(), key=lambda d: disc_space[d])
+                    episodes_per_disc[best_disc].append(ep)
+                    disc_space[best_disc] -= 10.0  # Rough estimate: reduce available space
+                    print(f"    ↻ Reassigning Episode {ep} to Disc {best_disc} (most available space)")
+        
         return episodes_per_disc
     
     def rename_episodes(self, episodes: List[Dict], tvdb_episodes: Optional[List[Dict]] = None) -> List[Dict]:
@@ -199,37 +374,136 @@ class EpisodeRenamer:
         return renamed_files
     
     def _match_episodes_by_duration(self, episodes: List[Dict], tvdb_episodes: List[Dict]) -> List[Dict]:
-        """Match episode files with TVDB episode data based on duration and lexicographical order."""
+        """Match episode files with TVDB episode data. Use disc mapping if provided, otherwise use duration matching."""
         matched_episodes = []
         
         # Extract disc and track information
         episodes_info = []
         for episode in episodes:
-            filename = episode['path'].name
+            # Handle both path objects and filename strings
+            if isinstance(episode, dict) and 'path' in episode:
+                filename = episode['path'].name
+            elif isinstance(episode, dict) and 'name' in episode:
+                filename = episode['name']
+            else:
+                print(f"Warning: Unexpected episode format: {episode}")
+                continue
+                
             disc_info = self._extract_disc_info(filename)
             episodes_info.append({
                 'file_info': episode,
                 'disc_number': disc_info['disc_number'],
                 'track_number': disc_info['track_number'],
                 'filename': filename,
-                'duration': episode['duration_minutes']
+                'duration': episode.get('duration_minutes', 0)
             })
         
         # Sort by lexicographical order (this should match episode order)
         episodes_info.sort(key=lambda x: x['filename'])
         
-        # If we have disc information, use it for context but prioritize duration matching
+        # Detect and handle duplicates before processing
+        print("\n=== Duplicate Detection ===")
+        episode_files = [info['file_info'] for info in episodes_info]
+        duplicate_files = self.detect_and_move_duplicates(episode_files)
+        
+        # Remove duplicates from episodes_info
+        if duplicate_files:
+            duplicate_names = {Path(f).name for f in duplicate_files}
+            episodes_info = [info for info in episodes_info 
+                           if info['filename'] not in duplicate_names]
+            print(f"Removed {len(duplicate_files)} duplicates from episode processing")
+        
+        # Use disc mapping if provided
+        if self.disc_mapping:
+            print("\n=== Using Disc Mapping ===")
+            matched_episodes = self._match_using_disc_mapping(episodes_info, tvdb_episodes)
+        else:
+            # Original complex matching logic for when no disc mapping is provided
+            print("\n=== Using Duration/Disc Analysis ===")
+            matched_episodes = self._match_using_duration_analysis(episodes_info, tvdb_episodes)
+        
+        # Sort by episode number for final output
+        matched_episodes.sort(key=lambda x: x['episode_number'])
+        return matched_episodes
+    
+    def _match_using_disc_mapping(self, episodes_info: List[Dict], tvdb_episodes: List[Dict]) -> List[Dict]:
+        """Match episodes using the provided disc mapping."""
+        matched_episodes = []
+        
+        # Group files by disc
+        disc_groups = {}
+        for info in episodes_info:
+            disc_num = info['disc_number']
+            if disc_num is not None:
+                disc_groups.setdefault(disc_num, []).append(info)
+        
+        print(f"Files found on {len(disc_groups)} discs")
+        for disc_num in sorted(disc_groups.keys()):
+            files = disc_groups[disc_num]
+            expected_episodes = self.disc_mapping.get(disc_num, [])
+            print(f"Disc {disc_num}: {len(files)} files, expecting episodes {expected_episodes}")
+            
+            # Sort files on this disc by filename
+            files.sort(key=lambda x: x['filename'])
+            
+            # Match files to episodes in order
+            for i, file_info in enumerate(files):
+                if i < len(expected_episodes):
+                    episode_number = expected_episodes[i]
+                    
+                    # Find the TVDB episode
+                    tvdb_episode = None
+                    for ep in tvdb_episodes:
+                        if ep['episode_number'] == episode_number:
+                            tvdb_episode = ep
+                            break
+                    
+                    matched_episodes.append({
+                        'file_info': file_info['file_info'],
+                        'episode_number': episode_number,
+                        'tvdb_info': tvdb_episode
+                    })
+                    
+                    print(f"  Episode {episode_number}: {file_info['filename']}")
+                else:
+                    print(f"  Warning: Extra file on disc {disc_num}: {file_info['filename']}")
+        
+        return matched_episodes
+    
+    def _match_using_duration_analysis(self, episodes_info: List[Dict], tvdb_episodes: List[Dict]) -> List[Dict]:
+        """Use the original complex duration/disc analysis matching."""
+        # Diagnostics: overview of discs/files and durations
+        try:
+            disc_groups = {}
+            for info in episodes_info:
+                disc_groups.setdefault(info['disc_number'], []).append(info)
+            print("=== Disc/File Overview ===")
+            for disc_key in sorted([d for d in disc_groups.keys() if d is not None]):
+                files = sorted(disc_groups[disc_key], key=lambda x: x['filename'])
+                print(f"Disc {disc_key}: {len(files)} candidate file(s)")
+                for f in files:
+                    try:
+                        print(f"  - {f['filename']} ({f['duration']:.1f} min)")
+                    except Exception:
+                        pass
+        except Exception:
+            pass
+        
+        # If we have disc information, enforce disc→episode ordering so early discs map to earlier episodes
         if any(info['disc_number'] for info in episodes_info):
-            print("Found disc information - using flexible duration-first matching...")
+            total_eps = len(tvdb_episodes)
-            # Use flexible matching that prioritizes duration over disc constraints
+            episodes_per_disc = self._estimate_episodes_per_disc(episodes_info, total_eps)
-            matched_episodes = self._flexible_duration_match(episodes_info, tvdb_episodes)
+            if episodes_per_disc:
+                print("Found disc information - enforcing disc-to-episode order constraints...")
+                matched_episodes = self._match_by_duration_and_disc(episodes_info, tvdb_episodes, episodes_per_disc)
+            else:
+                print("Found disc information - using flexible duration-first matching...")
+                matched_episodes = self._flexible_duration_match(episodes_info, tvdb_episodes)
         else:
             # No disc info, use lexicographical order with precise duration matching
             print("No disc information found, using lexicographical order with precise duration matching...")
             matched_episodes = self._match_by_duration_and_order(episodes_info, tvdb_episodes)
         
-        # Sort by episode number for final output
-        matched_episodes.sort(key=lambda x: x['episode_number'])
         return matched_episodes
     
     def _flexible_duration_match(self, episodes_info: List[Dict], tvdb_episodes: List[Dict]) -> List[Dict]:
@@ -492,7 +766,7 @@ class EpisodeRenamer:
         return True
 
     def _analyze_disc_capacity(self, episodes_info: List[Dict]) -> Dict[int, int]:
-        """Analyze the maximum number of episodes each disc can hold."""
+        """Analyze the maximum number of episodes each disc can hold based on storage constraints."""
         disc_capacity = {}
         
         # Group files by disc
@@ -504,9 +778,30 @@ class EpisodeRenamer:
                     disc_files[disc_num] = []
                 disc_files[disc_num].append(info)
         
-        # Each disc can hold at most as many episodes as it has files
+        # Analyze each disc's capacity considering storage limits
         for disc_num, files in disc_files.items():
-            disc_capacity[disc_num] = len(files)
+            file_count = len(files)
+            total_size_gb = sum(f['file_info']['size_gb'] for f in files)
+            
+            # Blu-ray capacity constraints
+            max_bluray_capacity_gb = 48.0
+            usable_capacity_gb = 45.0  # Account for overhead and extras
+            
+            if total_size_gb > usable_capacity_gb * 0.8:  # 80% threshold
+                # This disc is likely near capacity
+                # Estimate how many episodes can actually fit based on average episode size
+                avg_episode_size = total_size_gb / file_count if file_count > 0 else 10.0
+                estimated_episodes = min(file_count, int(usable_capacity_gb / avg_episode_size))
+                
+                # Be conservative - if we're unsure, use the smaller number
+                disc_capacity[disc_num] = min(estimated_episodes, file_count)
+                
+                print(f"  Storage constraint analysis for Disc {disc_num}:")
+                print(f"    Total files: {file_count}, Total size: {total_size_gb:.1f}GB")
+                print(f"    Estimated episode capacity: {estimated_episodes} (limited by {usable_capacity_gb}GB Blu-ray)")
+            else:
+                # Disc has plenty of space, use file count
+                disc_capacity[disc_num] = file_count
         
         return disc_capacity
     
@@ -553,11 +848,14 @@ class EpisodeRenamer:
     def _match_by_duration_and_disc(self, episodes_info: List[Dict], tvdb_episodes: List[Dict], episodes_per_disc: Dict[int, List[int]]) -> List[Dict]:
         """Match episodes using disc constraints and precise duration matching."""
         matched_episodes = []
-        used_episode_numbers = set()        # Process each disc in order
+        used_episode_numbers = set()
+        
+        # Process each disc in order
         for disc_num in sorted(episodes_per_disc.keys()):
             files_on_disc = [f for f in episodes_info if f['disc_number'] == disc_num]
             files_on_disc.sort(key=lambda x: x['filename'])
             allowed_episodes = episodes_per_disc[disc_num]
+            print(f"\nDisc {disc_num}: allowed episode numbers {allowed_episodes}")
             
             # Get TVDB episodes for this disc
             disc_tvdb_episodes = [ep for ep in tvdb_episodes if ep['episode_number'] in allowed_episodes]
@@ -609,6 +907,7 @@ class EpisodeRenamer:
             file_duration = file_info['duration']
             best_match = None
             best_score = float('inf')
+            diffs = []  # Diagnostics: collect per-episode duration differences
             
             for tvdb_ep in tvdb_episodes:
                 if tvdb_ep['episode_number'] in used_episodes:
@@ -622,6 +921,7 @@ class EpisodeRenamer:
                 
                 # Calculate duration difference
                 duration_diff = abs(file_duration - tvdb_duration)
+                diffs.append((tvdb_ep['episode_number'], tvdb_duration, duration_diff))
                 
                 if duration_diff < best_score:
                     best_score = duration_diff
@@ -637,6 +937,15 @@ class EpisodeRenamer:
                 })
                 used_files.add(id(file_info))
                 used_episodes.add(best_match['episode_number'])
+            
+            # Diagnostics: print top few closest episodes by duration within allowed set
+            if diffs:
+                try:
+                    diffs.sort(key=lambda x: x[2])
+                    top = ", ".join([f"E{ep}:{dur} (Δ{diff:.1f})" for ep, dur, diff in diffs[:3]])
+                    print(f"    {file_info['filename']}: compare within {allowed_episodes} → {top}")
+                except Exception:
+                    pass
         
         # Second pass: Handle remaining files with sequential assignment
         remaining_files = [f for f in files if id(f) not in used_files]

src/Matcher/file_classifier.py

@@ -144,91 +144,144 @@ class FileClassifier:
         # If we have TVDB episode data, use it for smarter classification
         if tvdb_episodes and len(tvdb_episodes) > 0:
             return self._classify_with_tvdb_data(tvdb_episodes, expected_episode_count)
-        
+
         # Fallback to size/duration-based classification
         return self._classify_by_stats(expected_episode_count)
-    
+
     def _classify_with_tvdb_data(self, tvdb_episodes: List[Dict], expected_episode_count: int = None) -> Tuple[List[Dict], List[Dict]]:
-        """Classify files using TVDB episode duration data for better matching."""
+        """TVDB-informed classification with disc-aware capacity and fair distribution."""
         if not tvdb_episodes:
             return self._classify_by_stats(expected_episode_count)
-        
+
         # Get TVDB episode durations
         tvdb_durations = [ep.get('runtime', 0) for ep in tvdb_episodes if ep.get('runtime', 0) > 0]
         if not tvdb_durations:
             return self._classify_by_stats(expected_episode_count)
-        
+
         avg_tvdb_duration = sum(tvdb_durations) / len(tvdb_durations)
         min_episode_duration = min(tvdb_durations) - 5  # 5 minute tolerance below minimum
         max_episode_duration = max(tvdb_durations) + 10  # 10 minute tolerance above maximum
-        
+
         print(f"TVDB Episode duration range: {min_episode_duration:.1f} - {max_episode_duration:.1f} min (avg: {avg_tvdb_duration:.1f})")
-        
+
-        episodes = []
+        # Step 1: Build candidate pool limited by duration range
-        extras = []
+        import re
-        
+        disc_regex = re.compile(r'(?i)\bdis[ck][\s._-]*(\d{1,3})(?!\d)')
-        # Sort files by size (largest first) for consistent processing
+
-        sorted_files = sorted(self.file_info, key=lambda x: x['size_bytes'], reverse=True)
+        candidates_by_disc = {}
-        
+        non_candidate_extras = []
-        for file_info in sorted_files:
+
+        for file_info in sorted(self.file_info, key=lambda x: x['size_bytes'], reverse=True):
             duration = file_info['duration_minutes']
-            is_extra = False
-            
-            # First check: Duration-based classification with tighter tolerances
             if duration < min_episode_duration or duration > max_episode_duration:
-                is_extra = True
+                non_candidate_extras.append(file_info)
                 print(f"  Duration-based extra: {file_info['name']} ({duration:.1f} min) - outside {min_episode_duration:.1f}-{max_episode_duration:.1f} range")
-            
+                continue
-            # Second check: If we have expected episode count, limit episodes (but be disc-aware)
+
-            elif expected_episode_count and len(episodes) >= expected_episode_count:
+            m = disc_regex.search(file_info['name'])
-                # Extract disc number for disc-aware logic
+            disc_num = int(m.group(1)) if m else None
-                import re
+            candidates_by_disc.setdefault(disc_num, []).append(file_info)
-                disc_match = re.search(r'[Dd]isc\s*(\d+)', file_info['name'], re.IGNORECASE)
+
-                file_disc = int(disc_match.group(1)) if disc_match else None
+        # If no expected count, fallback to previous simple path using candidates as episodes
-                
+        if not expected_episode_count:
-                # Check how many files from each disc we already have as episodes
+            episodes = [f for disc_files in candidates_by_disc.values() for f in disc_files]
-                disc_episode_counts = {}
+            episodes.sort(key=lambda x: x['name'])
-                for ep in episodes:
+            extras = non_candidate_extras
-                    ep_disc_match = re.search(r'[Dd]isc\s*(\d+)', ep['name'], re.IGNORECASE)
+            return episodes, extras
-                    ep_disc = int(ep_disc_match.group(1)) if ep_disc_match else 0
+
-                    disc_episode_counts[ep_disc] = disc_episode_counts.get(ep_disc, 0) + 1
+        # Step 2: Derive per-disc capacity cap from average runtime (no show-specific hardcoding)
-                
+        MAX_MINUTES_PER_DISC = 240  # Conservative, configurable heuristic
-                current_disc_episodes = disc_episode_counts.get(file_disc, 0)
+        capacity_per_disc = max(1, int(MAX_MINUTES_PER_DISC / avg_tvdb_duration))
-                
+
-                # Special logic: ensure we have enough episodes from Disc 4 for episodes 9 & 10
+        # Step 3: Fair-share distribution across discs with candidates
-                # Check if we need more Disc 4 episodes and this could be Episode 10
+        discs_in_play = sorted([d for d in candidates_by_disc.keys() if d is not None])
-                if file_disc == 4 and current_disc_episodes < 2:
+        if not discs_in_play and None in candidates_by_disc:
-                    print(f"  Disc 4 episode preserved: {file_info['name']} - ensuring final disc has episodes 9-10")
+            # No disc labels; treat as a single disc
-                    is_extra = False
+            discs_in_play = [None]
-                else:
+
-                    is_extra = True
+        n_discs = len(discs_in_play)
-                    print(f"  Count-based extra: {file_info['name']} (episode limit reached)")
+        base = expected_episode_count // n_discs if n_discs else expected_episode_count
-            
+        rem = expected_episode_count % n_discs if n_discs else 0
-            # Third check: Look for very close duration matches to confirm episodes
+
-            elif expected_episode_count:
+        # Initial caps per disc: fair share limited by capacity and available candidates
-                # Find closest TVDB episode duration
+        caps = {}
-                closest_duration = min(tvdb_durations, key=lambda x: abs(x - duration))
+        allocated = 0
-                duration_diff = abs(duration - closest_duration)
+        for idx, d in enumerate(discs_in_play):
-                
+            fair_cap = base + (1 if idx < rem else 0)
-                if duration_diff <= 10:  # Within 10 minutes is likely an episode
+            avail = len(candidates_by_disc.get(d, []))
-                    print(f"  Duration-matched episode: {file_info['name']} ({duration:.1f} min) ≈ TVDB {closest_duration} min (Δ{duration_diff:.1f})")
+            cap = min(fair_cap, capacity_per_disc, avail)
-                else:
+            caps[d] = cap
-                    # Check if this could still be an episode based on count
+            allocated += cap
-                    if len(episodes) < expected_episode_count:
+
-                        print(f"  Episode by count: {file_info['name']} ({duration:.1f} min) - no close TVDB match but within count")
+        # Redistribute any remaining quota without exceeding per-disc capacity or availability
-                    else:
+        remaining = expected_episode_count - allocated
-                        is_extra = True
+        if remaining > 0:
-                        print(f"  Duration mismatch extra: {file_info['name']} ({duration:.1f} min) - closest TVDB: {closest_duration} min (Δ{duration_diff:.1f})")
+            for _ in range(3):
-            
+                if remaining <= 0:
-            if is_extra:
+                    break
-                extras.append(file_info)
+                for d in discs_in_play:
-            else:
+                    if remaining <= 0:
-                episodes.append(file_info)
+                        break
-        
+                    avail = len(candidates_by_disc.get(d, []))
-        # Sort episodes by name for consistent ordering
+                    if caps[d] < min(capacity_per_disc, avail):
-        episodes.sort(key=lambda x: x['name'])
+                        caps[d] += 1
-        
+                        remaining -= 1
-        return episodes, extras
+
+        # Step 4: Select episodes per disc obeying caps and avoiding duplicates if over cap
+        def pick_for_disc(files: List[Dict], cap: int) -> List[Dict]:
+            if cap <= 0:
+                return []
+            if len(files) <= cap:
+                return sorted(files, key=lambda x: x['name'])
+
+            buckets = {}
+            for f in files:
+                key = round(f['duration_minutes'] * 2) / 2.0
+                buckets.setdefault(key, []).append(f)
+
+            for k in buckets:
+                buckets[k].sort(key=lambda x: (-x['size_bytes'], x['name']))
+
+            selected = []
+            bucket_keys = sorted(buckets.keys())
+            idx = 0
+            while len(selected) < cap and bucket_keys:
+                key = bucket_keys[idx % len(bucket_keys)]
+                if buckets[key]:
+                    selected.append(buckets[key].pop(0))
+                bucket_keys = [k for k in bucket_keys if buckets[k]]
+                idx += 1
+
+            if len(selected) < cap:
+                remaining_files = []
+                for arr in buckets.values():
+                    remaining_files.extend(arr)
+                remaining_files.sort(key=lambda x: (-x['size_bytes'], x['name']))
+                for f in remaining_files:
+                    if len(selected) >= cap:
+                        break
+                    selected.append(f)
+
+            return sorted(selected, key=lambda x: x['name'])
+
+        selected_episodes = []
+        for d in discs_in_play:
+            disc_files = candidates_by_disc.get(d, [])
+            chosen = pick_for_disc(disc_files, caps.get(d, 0))
+            for f in chosen:
+                closest = min(tvdb_durations, key=lambda x: abs(x - f['duration_minutes']))
+                print(f"  Duration-matched episode: {f['name']} ({f['duration_minutes']:.1f} min) ≈ TVDB {closest} min (Δ{abs(f['duration_minutes']-closest):.1f})")
+            selected_episodes.extend(chosen)
+
+        selected_set = {id(f) for f in selected_episodes}
+        extras = non_candidate_extras[:]
+        for files in candidates_by_disc.values():
+            for f in files:
+                if id(f) not in selected_set:
+                    extras.append(f)
+
+        selected_episodes.sort(key=lambda x: x['name'])
+        return selected_episodes, extras
     
     def _classify_by_stats(self, expected_episode_count: int = None) -> Tuple[List[Dict], List[Dict]]:
         """Fallback classification using statistical analysis of file sizes and durations."""

test_episode_matching.py

@@ -0,0 +1,185 @@
+#!/usr/bin/env python3
+"""
+Test script to verify episode matching logic with mock data.
+"""
+import sys
+import os
+from pathlib import Path
+
+# Add src to path
+sys.path.insert(0, os.path.join(os.path.dirname(__file__), 'src'))
+
+from Matcher.episode_renamer import EpisodeRenamer
+
+def create_mock_episodes():
+    """Create mock episode data to test the matching logic."""
+    return [
+        {
+            'path': Path('Disc 1_t01.mkv'),
+            'name': 'Disc 1_t01.mkv',
+            'size_bytes': 2 * 1024**3,  # 2GB
+            'size_gb': 2.0,
+            'duration_minutes': 61.0,
+        },
+        {
+            'path': Path('Disc 1_t02.mkv'), 
+            'name': 'Disc 1_t02.mkv',
+            'size_bytes': 2.1 * 1024**3,  # 2.1GB
+            'size_gb': 2.1,
+            'duration_minutes': 55.0,
+        },
+        {
+            'path': Path('Disc 1_t03.mkv'),
+            'name': 'Disc 1_t03.mkv', 
+            'size_bytes': 2.2 * 1024**3,  # 2.2GB
+            'size_gb': 2.2,
+            'duration_minutes': 57.0,
+        },
+        {
+            'path': Path('Disc 2_t01.mkv'),
+            'name': 'Disc 2_t01.mkv',
+            'size_bytes': 2.0 * 1024**3,  # 2GB
+            'size_gb': 2.0,
+            'duration_minutes': 55.0,
+        },
+        {
+            'path': Path('Disc 2_t02.mkv'),
+            'name': 'Disc 2_t02.mkv',
+            'size_bytes': 2.1 * 1024**3,  # 2.1GB  
+            'size_gb': 2.1,
+            'duration_minutes': 54.0,
+        },
+        {
+            'path': Path('Disc 2_t03.mkv'),
+            'name': 'Disc 2_t03.mkv',
+            'size_bytes': 1.9 * 1024**3,  # 1.9GB
+            'size_gb': 1.9,
+            'duration_minutes': 52.0,
+        },
+        {
+            'path': Path('Disc 3_t01.mkv'),
+            'name': 'Disc 3_t01.mkv',
+            'size_bytes': 2.2 * 1024**3,  # 2.2GB
+            'size_gb': 2.2,
+            'duration_minutes': 57.0,
+        },
+        {
+            'path': Path('Disc 3_t02.mkv'),
+            'name': 'Disc 3_t02.mkv', 
+            'size_bytes': 2.1 * 1024**3,  # 2.1GB
+            'size_gb': 2.1,
+            'duration_minutes': 58.0,
+        }
+    ]
+
+def create_mock_tvdb_episodes():
+    """Create mock TVDB episode data."""
+    return [
+        {'episode_number': 1, 'name': 'Winter Is Coming', 'runtime': 61},
+        {'episode_number': 2, 'name': 'The Kingsroad', 'runtime': 55},
+        {'episode_number': 3, 'name': 'Lord Snow', 'runtime': 57},
+        {'episode_number': 4, 'name': 'Cripples, Bastards, and Broken Things', 'runtime': 55},
+        {'episode_number': 5, 'name': 'The Wolf and the Lion', 'runtime': 54},
+        {'episode_number': 6, 'name': 'A Golden Crown', 'runtime': 52},
+        {'episode_number': 7, 'name': 'You Win or You Die', 'runtime': 57},
+        {'episode_number': 8, 'name': 'The Pointy End', 'runtime': 58},
+        {'episode_number': 9, 'name': 'Baelor', 'runtime': 56},
+        {'episode_number': 10, 'name': 'Fire and Blood', 'runtime': 52},
+    ]
+
+def test_episode_matching():
+    """Test the episode matching logic."""
+    print("=== Testing Episode Matching Logic ===\n")
+    
+    # Create mock data
+    episodes = create_mock_episodes()
+    tvdb_episodes = create_mock_tvdb_episodes()
+    
+    print(f"Mock episode files ({len(episodes)}):")
+    for ep in episodes:
+        print(f"  - {ep['name']} ({ep['duration_minutes']} min)")
+    
+    print(f"\nTVDB episodes ({len(tvdb_episodes)}):")
+    for ep in tvdb_episodes:
+        print(f"  - Episode {ep['episode_number']}: {ep['name']} ({ep['runtime']} min)")
+    
+    # Initialize renamer
+    renamer = EpisodeRenamer("/tmp/test", "Game of Thrones", 1)
+    
+    # Test the matching logic
+    print("\n" + "="*80)
+    print("TESTING EPISODE MATCHING")
+    print("="*80)
+    
+    try:
+        matched_episodes = renamer._match_episodes_by_duration(episodes, tvdb_episodes)
+        
+        print(f"\n=== MATCHING RESULTS ===")
+        print(f"Successfully matched {len(matched_episodes)} episodes:")
+        
+        for match in matched_episodes:
+            file_name = match['file_info']['path'].name
+            episode_num = match['episode_number']
+            tvdb_name = match['tvdb_info']['name'] if match['tvdb_info'] else 'Unknown'
+            duration_diff = match.get('duration_diff', 'N/A')
+            
+            print(f"  {file_name} → Episode {episode_num}: {tvdb_name}")
+            if duration_diff != 'N/A' and duration_diff is not None:
+                print(f"    Duration difference: {duration_diff:.1f} minutes")
+        
+        # Test validity of assignments
+        print(f"\n=== ASSIGNMENT VALIDATION ===")
+        episode_numbers = [m['episode_number'] for m in matched_episodes]
+        if len(set(episode_numbers)) == len(episode_numbers):
+            print("✓ All episodes assigned unique episode numbers")
+        else:
+            print("✗ Duplicate episode number assignments detected!")
+            duplicates = [x for x in episode_numbers if episode_numbers.count(x) > 1]
+            print(f"  Duplicates: {set(duplicates)}")
+        
+        # Check disc order constraint 
+        disc_assignments = {}
+        for match in matched_episodes:
+            file_name = match['file_info']['path'].name
+            episode_num = match['episode_number']
+            
+            # Extract disc number
+            if 'Disc 1' in file_name:
+                disc = 1
+            elif 'Disc 2' in file_name:
+                disc = 2
+            elif 'Disc 3' in file_name:
+                disc = 3
+            else:
+                disc = None
+                
+            if disc:
+                if disc not in disc_assignments:
+                    disc_assignments[disc] = []
+                disc_assignments[disc].append(episode_num)
+        
+        print(f"\nDisc assignment validation:")
+        for disc in sorted(disc_assignments.keys()):
+            episodes = sorted(disc_assignments[disc])
+            print(f"  Disc {disc}: Episodes {episodes}")
+            
+            # Check if episodes are in reasonable order (earlier discs should have earlier episodes)
+            if disc == 1:
+                expected_range = list(range(1, 4))  # Episodes 1-3
+            elif disc == 2: 
+                expected_range = list(range(4, 7))  # Episodes 4-6
+            elif disc == 3:
+                expected_range = list(range(7, 9))  # Episodes 7-8
+            
+            if set(episodes).issubset(set(expected_range)):
+                print(f"    ✓ Episodes {episodes} are within expected range {expected_range}")
+            else:
+                print(f"    ⚠ Episodes {episodes} outside expected range {expected_range}")
+    
+    except Exception as e:
+        print(f"Error during matching: {e}")
+        import traceback
+        traceback.print_exc()
+
+if __name__ == "__main__":
+    test_episode_matching()