pdf?

app.py

@@ -42,17 +42,46 @@ processor = Qwen2VLProcessor.from_pretrained(model_id)
 
 def process_uploaded_file(file_obj):
     """Process uploaded file whether it's an image or PDF"""
-    # Get the file extension
     file_extension = Path(file_obj.name).suffix.lower()
     
     try:
         if file_extension == '.pdf':
-            # Convert PDF to images
-            images = pdf2image.convert_from_path(file_obj.name)
-            return [img for img in images]
+            try:
+                # For PDF files, we need to use a temporary directory
+                with tempfile.TemporaryDirectory() as temp_dir:
+                    temp_pdf_path = os.path.join(temp_dir, "temp.pdf")
+                    
+                    # Save the uploaded PDF to the temporary path
+                    with open(file_obj.name, 'rb') as src_file:
+                        with open(temp_pdf_path, 'wb') as dst_file:
+                            dst_file.write(src_file.read())
+                    
+                    # Convert PDF to images using pdf2image
+                    try:
+                        images = pdf2image.convert_from_path(
+                            temp_pdf_path,
+                            poppler_path=None,  # Will use system poppler if available
+                            dpi=200,  # Adjust DPI as needed
+                            fmt='PNG'
+                        )
+                        return images
+                    except Exception as pdf_error:
+                        if "poppler" in str(pdf_error).lower():
+                            raise Exception(
+                                "PDF processing requires poppler to be installed. "
+                                "Please install poppler-utils package on your system. "
+                                "On Ubuntu/Debian: sudo apt-get install -y poppler-utils"
+                            )
+                        raise
+            except Exception as e:
+                raise Exception(f"PDF processing error: {str(e)}")
         else:
             # Handle regular image files
-            return [Image.open(file_obj.name)]
+            try:
+                img = Image.open(file_obj.name)
+                return [img]
+            except Exception as e:
+                raise Exception(f"Image processing error: {str(e)}")
     except Exception as e:
         raise Exception(f"Error processing file {file_obj.name}: {str(e)}")
 
@@ -78,21 +107,26 @@ def run_example(files, text_input=None):
         for file in files:
             try:
                 images = process_uploaded_file(file)
-                processed_images.extend(images)
+                if images:  # Only add if we got valid images
+                    processed_images.extend(images)
             except Exception as e:
                 return f"Error processing file {file.name}: {str(e)}"
         
         if not processed_images:
-            return "No valid images were processed. Please check your files."
+            return "No valid images were processed. Please check your files and ensure they are valid image or PDF files."
             
         # Save processed images temporarily
         image_paths = []
         for idx, img in enumerate(processed_images):
-            timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
-            filename = f"temp_image_{timestamp}_{idx}.png"
-            img.save(filename)
-            image_paths.append(filename)
-            temp_paths.append(filename)
+            try:
+                timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
+                filename = f"temp_image_{timestamp}_{idx}.png"
+                img.save(filename)
+                image_paths.append(filename)
+                temp_paths.append(filename)
+            except Exception as e:
+                cleanup_temp_files(temp_paths)  # Clean up any files we've created so far
+                return f"Error saving processed image: {str(e)}"
         
         # Create messages with multiple images
         messages = [

requirements.txt

@@ -7,6 +7,6 @@ git+https://github.com/huggingface/transformers.git
 accelerate
 qwen-vl-utils
 peft
-pdf2image
-pypdf2
+pdf2image>=1.16.3
+pypdf2>=3.0.0
 gradio>=4.0.0