Faster grouping, in order, with threads

grouping.py

@@ -4,6 +4,7 @@ import re
 import sys
 import shutil
 from collections import defaultdict
+from concurrent.futures import ThreadPoolExecutor
 from PIL import Image, ImageDraw, ImageFont
 from pdf2image import convert_from_path, pdfinfo_from_path
 
@@ -105,40 +106,6 @@ def group_files(file_list):
     # Return list of lists (strip the metadata)
     return [g['items'] for g in groups]
 
-# def group_files(file_list):
-#     """Groups files based on constraints."""
-#     sorted_files = sorted(file_list, key=lambda x: x[0])
-
-#     groups = []
-#     current_group = []
-#     current_height = 0
-
-#     for item in sorted_files:
-#         dd, path, height = item
-
-#         # Calculate added height (image + separator + approx text space)
-#         # We add separator height only if it's not the first image
-#         added_overhead = SEPARATOR_HEIGHT + 30 if current_group else 0
-
-#         # Check conditions
-#         if (len(current_group) >= MAX_GROUP_COUNT or
-#             (current_height + height + added_overhead) > MAX_GROUP_HEIGHT):
-
-#             # Push current group and start new
-#             if current_group:
-#                 groups.append(current_group)
-#             current_group = []
-#             current_height = 0
-#             added_overhead = 0 # Reset for first file of new group
-
-#         current_group.append(item)
-#         current_height += height + added_overhead
-
-#     if current_group:
-#         groups.append(current_group)
-
-#     return groups
-
 def stitch_pdf_pages(images_list):
     """Vertically concatenates a list of PIL images with no separator."""
     if not images_list:
@@ -242,6 +209,23 @@ def create_jpg(identifier, group_index, group, root_dir):
 
     print(f"Saved {output_path} with {len(group)} ({os.path.getsize(output_path)/1024/1024:.2f} MB)")
 
+def natural_key(text):
+    return [int(c) if c.isdigit() else c.lower() for c in re.split(r'(\d+)', str(text))]
+
+
+def process_identifier(identifier, files_info, root_dir):
+    # Clear output directory if it exists
+    target_folder = os.path.join(root_dir, identifier)
+    if os.path.exists(target_folder):
+        shutil.rmtree(target_folder)
+    os.makedirs(target_folder, exist_ok=True)
+
+    # files_info is list of (dd, path, height)
+    file_groups = group_files(files_info)
+
+    for idx, group in enumerate(file_groups):
+        create_jpg(identifier, idx, group, root_dir)
+
 def main():
     if len(sys.argv) < 2:
         print("Usage: python app.py <Path_to_Dir>")
@@ -254,18 +238,13 @@ def main():
 
     print(f"Found {len(data)} identifiers. Processing...")
 
-    for identifier, files_info in data.items():
-        # Clear output directory if it exists
-        target_folder = os.path.join(root_dir, identifier)
-        if os.path.exists(target_folder):
-            shutil.rmtree(target_folder)
-        os.makedirs(target_folder, exist_ok=True)
+    # Sort identifiers naturally
+    sorted_identifiers = sorted(data.keys(), key=natural_key)
 
-        # files_info is list of (dd, path, height)
-        file_groups = group_files(files_info)
-
-        for idx, group in enumerate(file_groups):
-            create_jpg(identifier, idx, group, root_dir)
+    # Process using 4 threads
+    with ThreadPoolExecutor(max_workers=4) as executor:
+        for identifier in sorted_identifiers:
+            executor.submit(process_identifier, identifier, data[identifier], root_dir)
 
     print("Done.")