Add support for de-duplicating chapters and selecting the best version

cbz_volume_combiner/volume.py

@@ -29,6 +29,55 @@ def create_volume_cbz(manga_name, volume_num, chapter_infos, output_dir=None, fo
         return True, "Skipped (already exists)"
     
     try:
+        # De-duplicate chapters (keep only one copy of each chapter number)
+        if len(chapter_infos) > 1:
+            # Group chapters by their chapter number
+            chapter_groups = {}
+            for chapter in chapter_infos:
+                chapter_num = str(chapter['chapter'])  # Convert to string for exact matching
+                if chapter_num not in chapter_groups:
+                    chapter_groups[chapter_num] = []
+                chapter_groups[chapter_num].append(chapter)
+            
+            # Check for duplicates
+            duplicates_found = False
+            for chapter_num, group in chapter_groups.items():
+                if len(group) > 1:
+                    duplicates_found = True
+                    break
+            
+            if duplicates_found:
+                # Create a new de-duplicated chapter list
+                unique_chapters = []
+                
+                if verbose:
+                    print("Found duplicate chapters, selecting one copy of each:")
+                
+                for chapter_num, group in chapter_groups.items():
+                    if len(group) > 1:
+                        # We have duplicates for this chapter
+                        if verbose:
+                            print(f"  Chapter {chapter_num} has {len(group)} copies:")
+                            for i, ch in enumerate(group):
+                                print(f"    {i+1}. {os.path.basename(ch['filename'])}")
+                        
+                        # Select the best version - prefer certain scan groups or larger files
+                        selected_chapter = select_best_chapter(group, extra_verbose)
+                        
+                        if verbose:
+                            print(f"  Selected: {os.path.basename(selected_chapter['filename'])}")
+                        
+                        unique_chapters.append(selected_chapter)
+                    else:
+                        # Only one copy, just add it
+                        unique_chapters.append(group[0])
+                
+                # Replace the original chapter list with the de-duplicated one
+                chapter_infos = unique_chapters
+                
+                if verbose:
+                    print(f"De-duplicated chapter list now has {len(chapter_infos)} chapters")
+        
         # Verify all chapter files exist before starting and attempt to fix missing files
         missing_files = []
         for chapter_info in chapter_infos:
@@ -177,4 +226,31 @@ def create_volume_cbz(manga_name, volume_num, chapter_infos, output_dir=None, fo
         
         return True, "Created"
     except Exception as e:
-        return False, str(e)
+        return False, str(e)
+
+def select_best_chapter(chapter_group, extra_verbose=False):
+    """
+    Select the best chapter from a group of duplicate chapters.
+    Strategy:
+    1. If all files exist, pick the largest one (likely better quality)
+    2. If some don't exist, pick one that exists
+    3. If none exist, just return the first one
+    """
+    existing_chapters = [ch for ch in chapter_group if os.path.exists(ch['filename'])]
+    
+    if not existing_chapters:
+        if extra_verbose:
+            print("Warning: None of the duplicate chapters exist on disk")
+        return chapter_group[0]  # Return the first one and hope for the best
+    
+    # Get file sizes for existing chapters
+    for chapter in existing_chapters:
+        chapter['filesize'] = os.path.getsize(chapter['filename'])
+    
+    # Sort by file size (descending) - bigger files are often better quality
+    existing_chapters.sort(key=lambda x: x['filesize'], reverse=True)
+    
+    if extra_verbose:
+        print(f"Selected largest file: {os.path.basename(existing_chapters[0]['filename'])} ({existing_chapters[0]['filesize'] / 1024:.1f} KB)")
+    
+    return existing_chapters[0]