První nahrdánícd

PDF optimization.py

@@ -0,0 +1,73 @@
+import io
+
+import PyPDF2
+import os
+
+from PyPDF2 import PdfReader, PdfWriter
+
+
+def optimize_pdf_with_images(input_file, destination_folder):
+    # """ Optimize a PDF by reducing the size of embedded images while preserving all other content. """/
+
+    optimized_pdf_path = os.path.join(destination_folder, os.path.basename(input_file))
+
+    pdf_reader = PdfReader(input_file)
+    pdf_writer = PdfWriter()
+
+    for pdf_page in pdf_reader.pages:
+        page_resources = pdf_page.get("/Resources")
+        if page_resources:
+            page_resources = page_resources.get_object()  # Resolve IndirectObject
+
+        if "/XObject" in page_resources:
+            image_objects = page_resources["/XObject"].get_object()
+
+            for img_name in image_objects:
+                img_obj = image_objects[img_name]
+
+                if img_obj.get("/Subtype") == "/Image":
+
+                    # Extract the image data
+                    image_data = img_obj._data
+                    image_stream = io.BytesIO(image_data)
+
+                    # Compress the image
+                    compressed_image_stream = compress_image_for_pdf(image_stream)
+
+                    if compressed_image_stream is None:
+                        print(f"Skipping invalid image: {img_name}")
+                        continue  # Skip this image if compression failed
+
+                    # Get new image dimensions
+                    resized_image = Image.open(compressed_image_stream)
+                    new_width, new_height = resized_image.size
+
+                    # Create a new image object with compressed data
+                    new_image_object = StreamObject()
+                    new_image_object._data = compressed_image_stream.getvalue()
+                    new_image_object.update({
+                        NameObject("/Filter"): NameObject("/DCTDecode"),
+                        NameObject("/Subtype"): NameObject("/Image"),
+                        NameObject("/Width"): NumberObject(new_width),
+                        NameObject("/Height"): NumberObject(new_height),
+                        NameObject("/ColorSpace"): NameObject("/DeviceRGB"),
+                        NameObject("/BitsPerComponent"): NumberObject(8),
+                    })
+
+                    # Replace the old image with the new one
+                    image_objects[img_name] = new_image_object
+
+        pdf_writer.add_page(pdf_page)
+
+    pdf_reader.stream.close()
+
+    print(f"PDF reader closed, now writing the optimized file.")
+    print(f"optimized_pdf_path : {optimized_pdf_path}")  # Log the file path
+    with open(optimized_pdf_path, "wb") as output_pdf:
+        pdf_writer.write(output_pdf)
+
+    print(f"Optimized PDF saved at: {optimized_pdf_path}")
+    return optimized_pdf_path
+
+optimize_pdf_with_images(r"u:\Dropbox\!!!Days\Downloads Z230\Dokumentace\6504140489 2025-03-04 Harsa, Radovan [PZ psychiatrie] [28FEB2025-04MAR2025 vágní suicidální proklamace, zbytečná hospitalizace].pdf",
+                         r"u:\Dropbox\!!!Days\Downloads Z230\Dokumentace\Converted")