Update app.py

app.py

@@ -4,13 +4,7 @@ import requests
 import io
 import tempfile
 import sys
-from PIL import Image
-
-def compress_image(image, quality):
-    img_buffer = io.BytesIO()
-    image.save(img_buffer, format='JPEG', quality=quality)
-    img_buffer.seek(0)
-    return img_buffer
+import os
 
 def compress_pdf(input_file, url, strength):
     if input_file is None and (url is None or url.strip() == ""):
@@ -30,53 +24,64 @@ def compress_pdf(input_file, url, strength):
         pdf_content = input_file
 
     try:
+        # Get the initial file size
+        pdf_content.seek(0, os.SEEK_END)
+        initial_size = pdf_content.tell()
+        pdf_content.seek(0)
+
         reader = PyPDF2.PdfReader(pdf_content)
         writer = PyPDF2.PdfWriter()
 
         if strength == "Low":
-            image_quality = 65
-            compression_level = 1
+            target_ratio = 0.75  # 25% compression
         elif strength == "Medium":
-            image_quality = 40
-            compression_level = 2
+            target_ratio = 0.50  # 50% compression
         else:  # High
-            image_quality = 20
-            compression_level = 3
+            target_ratio = 0.25  # 75% compression
 
+        # First pass: apply basic compression
         for page in reader.pages:
             page.compress_content_streams()  # Apply content stream compression
-            
-            # Compress images on the page
-            for img_index, img in enumerate(page.images):
-                if img.image is not None:
-                    try:
-                        pil_image = Image.open(io.BytesIO(img.image))
-                        compressed_image = compress_image(pil_image, image_quality)
-                        page.replace_image(img_index, compressed_image)
-                    except Exception as e:
-                        print(f"Error compressing image: {e}")
-
             writer.add_page(page)
 
+        # Set compression parameters
+        writer.compress = True
+        writer._compress_streams = True
+        writer._compress_pages = True
+
+        # Write the compressed PDF to a temporary file
         with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as temp_file:
             writer.write(temp_file)
             temp_file_path = temp_file.name
 
-        # Apply additional compression using PyPDF2's built-in compression
-        with open(temp_file_path, 'rb') as file:
-            reader = PyPDF2.PdfReader(file)
+        # Check the compression ratio achieved
+        compressed_size = os.path.getsize(temp_file_path)
+        current_ratio = compressed_size / initial_size
+
+        # If we haven't reached the target ratio, apply additional compression
+        if current_ratio > target_ratio:
+            reader = PyPDF2.PdfReader(temp_file_path)
             writer = PyPDF2.PdfWriter()
-            
+
             for page in reader.pages:
+                # Apply more aggressive compression
                 page.compress_content_streams()
                 writer.add_page(page)
-            
-            writer.add_metadata(reader.metadata)
-            
-            with open(temp_file_path, 'wb') as output_file:
-                writer.write(output_file)
 
-        return temp_file_path, "PDF compressed successfully!"
+            writer.compress = True
+            writer._compress_streams = True
+            writer._compress_pages = True
+
+            # Overwrite the temporary file with the more compressed version
+            with open(temp_file_path, 'wb') as temp_file:
+                writer.write(temp_file)
+
+        # Final compression ratio
+        final_size = os.path.getsize(temp_file_path)
+        final_ratio = final_size / initial_size
+        compression_percentage = (1 - final_ratio) * 100
+
+        return temp_file_path, f"PDF compressed successfully! Compression achieved: {compression_percentage:.2f}%"
     except Exception as e:
         return None, f"Error compressing PDF: {str(e)}"