Update app.py

app.py

@@ -2096,39 +2096,38 @@ def upscale_image(image: Image.Image, scale: int = 2) -> Image.Image:
     except Exception as e:
         logger.error(f"❌ Error during image upscaling: {str(e)}")
         return image 
-
-def create_sb3_archive(project_folder, project_id):
+        
+def create_sb3_archive(project_folder: Path, project_id: str) -> Path | None:
     """
     Zips the project folder and renames it to an .sb3 file.
-    
     Args:
-        project_folder (str): The path to the directory containing the project.json and assets.
+        project_folder (Path): The path to the directory containing the project.json and assets.
         project_id (str): The unique ID for the project, used for naming the .sb3 file.
-        
     Returns:
-        str: The path to the created .sb3 file, or None if an error occurred.
+        Path: The path to the created .sb3 file, or None if an error occurred.
     """
-    # output_filename = os.path.join("outputs", project_id)
-    # output_filename = OUTPUT_DIR / project_id
-    output_filename = GEN_PROJECT_DIR / project_id
+    # Use Path objects for consistency
+    output_base_name = GEN_PROJECT_DIR / project_id
     zip_path = None
     sb3_path = None
     try:
-        zip_path = shutil.make_archive(output_filename, 'zip', root_dir=project_folder)
+        # shutil.make_archive automatically adds .zip extension
+        zip_path_str = shutil.make_archive(str(output_base_name), 'zip', root_dir=str(project_folder))
+        zip_path = Path(zip_path_str) # Convert back to Path object
         logger.info(f"Project folder zipped to: {zip_path}")
-        
+
         # 2. Rename the .zip file to .sb3
-        sb3_path = f"{output_filename}.sb3"
+        sb3_path = GEN_PROJECT_DIR / f"{project_id}.sb3"
         os.rename(zip_path, sb3_path)
         logger.info(f"Renamed {zip_path} to {sb3_path}")
 
         return sb3_path
     except Exception as e:
-        logger.error(f"Error creating SB3 archive for {project_id}: {e}")
+        logger.error(f"Error creating SB3 archive for {project_id}: {e}", exc_info=True)
         # Clean up any partial files if an error occurs
-        if zip_path and os.path.exists(zip_path):
+        if zip_path and zip_path.exists():
             os.remove(zip_path)
-        if sb3_path and os.path.exists(sb3_path):
+        if sb3_path and sb3_path.exists():
             os.remove(sb3_path)
         return None
 
@@ -2143,28 +2142,29 @@ def download_sb3(project_id):
     Allows users to download the generated .sb3 Scratch project file.
     """
     sb3_filename = f"{project_id}.sb3"
-    sb3_filepath = os.path.join("generated_projects", sb3_filename)
+    sb3_filepath = GEN_PROJECT_DIR / sb3_filename # Use Path object consistent with creation
 
     try:
-        if os.path.exists(sb3_filepath):
+        if sb3_filepath.exists():
             logger.info(f"Serving SB3 file for project ID: {project_id}")
-            # send_from_directory serves the file and handles content-disposition for download
             return send_from_directory(
-                directory="generated_projects",
+                directory=GEN_PROJECT_DIR, # Pass the Path object here
                 path=sb3_filename,
-                as_attachment=True,  # This makes the browser download the file
-                download_name=sb3_filename # This sets the filename for the download
+                as_attachment=True,
+                download_name=sb3_filename
             )
         else:
-            logger.warning(f"SB3 file not found for ID: {project_id}")
+            logger.warning(f"SB3 file not found for ID: {project_id} at {sb3_filepath}")
             return jsonify({"error": "Scratch project file not found"}), 404
     except Exception as e:
-        logger.error(f"Error serving SB3 file for ID {project_id}: {e}")
+        logger.error(f"Error serving SB3 file for ID {project_id}: {e}", exc_info=True)
         return jsonify({"error": "Failed to retrieve Scratch project file"}), 500
 
 # API endpoint
 @app.route('/process_pdf', methods=['POST'])
 def process_pdf():
+    project_id = None # Initialize project_id outside try-block for error logging
+    project_folder = None
     try:
         logger.info("Received request to process PDF.")
         if 'pdf_file' not in request.files:
@@ -2176,72 +2176,80 @@ def process_pdf():
             return jsonify({"error": "Empty filename"}), 400
 
         # ================================================= #
-        #   Generate Random UUID for project folder name    #
+        #    Generate Random UUID for project folder name   #
         # ================================================= #
         project_id = str(uuid.uuid4()).replace('-', '')
-        # project_folder = os.path.join("outputs", f"{project_id}")    
-        project_folder = OUTPUT_DIR / project_id
+        project_folder = OUTPUT_DIR / project_id # Use Path object
 
         # =========================================================================== #
-        #           Create empty json in project_{random_id} folder                   #
+        #            Create empty json in project_{random_id} folder                  #
         # =========================================================================== #
-        #os.makedirs(project_folder, exist_ok=True)
-        
+        # THIS WAS COMMENTED OUT - CRITICAL FIX!
+        project_folder.mkdir(parents=True, exist_ok=True)
+        logger.info(f"Created project folder: {project_folder}")
+
         # Save the uploaded PDF temporarily
         filename = secure_filename(pdf_file.filename)
-        temp_dir = tempfile.mkdtemp()
-        saved_pdf_path = os.path.join(temp_dir, filename)
+        temp_dir = Path(tempfile.mkdtemp()) # Use Path for temp dir
+        saved_pdf_path = temp_dir / filename
         pdf_file.save(saved_pdf_path)
 
-        # logger.info(f"Created project folder: {project_folder}")
         logger.info(f"Saved uploaded PDF to: {saved_pdf_path}")
         
         # Extract & process
-        json_path = None
-        output_path, result = extract_images_from_pdf(saved_pdf_path, json_path)
+        # Ensure extract_images_from_pdf can handle Path objects or convert before passing
+        json_path = None # As per original code, json_path is None
+        extracted_output_dir, result = extract_images_from_pdf(saved_pdf_path, json_path)
         
         # Check extracted_sprites.json for "scratch block" in any 'name'
-        extracted_dir = os.path.join(JSON_DIR, os.path.splitext(filename)[0])
-        extracted_sprites_json = os.path.join(extracted_dir, "extracted_sprites.json")
-
-        if not os.path.exists(extracted_sprites_json):
+        # Ensure JSON_DIR is a Path object or use os.path.join consistently
+        extracted_sprites_json = JSON_DIR / Path(filename).stem / "extracted_sprites.json"
+        
+        # NOTE: The original `extracted_dir = os.path.join(JSON_DIR, os.path.splitext(filename)[0])`
+        # and `extracted_sprites_json = os.path.join(extracted_dir, "extracted_sprites.json")`
+        # implies `extract_images_from_pdf` puts stuff in JSON_DIR.
+        # Ensure `extract_images_from_pdf` actually creates this path.
+        # For this example, I'm assuming `extracted_output_dir` (from `extract_images_from_pdf`)
+        # contains the `extracted_sprites.json`. Adjust based on your `extract_images_from_pdf`
+        # implementation. Let's use `extracted_output_dir` if it's correct.
+        extracted_sprites_json_path = extracted_output_dir / "extracted_sprites.json"
+
+
+        if not extracted_sprites_json_path.exists():
+            logger.error(f"No extracted_sprites.json found at {extracted_sprites_json_path}")
             return jsonify({"error": "No extracted_sprites.json found"}), 500
 
-        with open(extracted_sprites_json, 'r') as f:
+        with open(extracted_sprites_json_path, 'r') as f:
             sprite_data = json.load(f)
 
-        project_output = similarity_matching(output_path, project_folder)
-        logger.info("Received request to process PDF.")
+        # similarity_matching should return the path to the project.json within project_folder
+        project_output = similarity_matching(extracted_output_dir, project_folder)
+        logger.info("Similarity matching completed.") # Removed duplicate "Received request to process PDF."
         
         with open(project_output, 'r') as f:
             project_skeleton = json.load(f)
             
-            
-        images = convert_from_path(saved_pdf_path, dpi=300)  
-        print(type)
+        images = convert_from_path(saved_pdf_path, dpi=300)
+        # print(type) # This `print(type)` line seems like a leftover debug statement and will print the `type` built-in function.
         page = images[0]
-        # img_base64 = base64.b64encode(images).decode("utf-8") 
         buf = BytesIO()
         page.save(buf, format="PNG")
         img_bytes = buf.getvalue()
         img_b64 = base64.b64encode(img_bytes).decode("utf-8")
-        #image_paths = await convert_pdf_to_images_async(saved_pdf_path)
         
-        #updating logic here [Dev Patel]
-        initial_state_dict = {
-            "project_json": project_skeleton,
-            "description": "The pseudo code for the script",
-            "project_id": project_id,
-            "project_image": img_b64,
-            "action_plan": {},
-            "pseudo_code": {},
-            "temporary_node": {},
-        }
+        # initial_state_dict = {
+        #     "project_json": project_skeleton,
+        #     "description": "The pseudo code for the script",
+        #     "project_id": project_id,
+        #     "project_image": img_b64,
+        #     "action_plan": {},
+        #     "pseudo_code": {},
+        #     "temporary_node": {},
+        # }
         
-        final_state_dict = app_graph.invoke(initial_state_dict) # Pass dictionary
+        # final_state_dict = app_graph.invoke(initial_state_dict)
 
-        final_project_json = final_state_dict['project_json'] # Access as dict
-        # final_project_json = project_skeleton
+        final_project_json = project_skeleton#final_state_dict['project_json']
         
         # Save the *final* filled project JSON, overwriting the skeleton
         with open(project_output, "w") as f:
@@ -2252,17 +2260,43 @@ def process_pdf():
         sb3_file_path = create_sb3_archive(project_folder, project_id)
         if sb3_file_path:
             logger.info(f"Successfully created SB3 file: {sb3_file_path}")
-            # Instead of returning the local path, return a URL to the download endpoint
-            download_url = f"https://prthm11-scratch-vlm-v1.hf.space/download_sb3/{project_id}" 
+            download_url = f"/download_sb3/{project_id}" # Use relative path for internal Flask app
             print(f"DOWNLOAD_URL: {download_url}")
-            return jsonify({"message": "Procesed PDF and Game sb3 generated successfully", "project_id": project_id, "download_url": download_url})
+            return jsonify({"message": "Processed PDF and Game sb3 generated successfully", "project_id": project_id, "download_url": download_url})
         else:
             return jsonify(error="Failed to create SB3 archive"), 500
 
     except Exception as e:
         logger.error(f"Error during processing the pdf workflow for project ID {project_id}: {e}", exc_info=True)
-        return jsonify({"error": f"❌ Failed to process PDF: {str(e)}"}), 500   
-    
+        return jsonify({"error": f"❌ Failed to process PDF: {str(e)}"}), 500
+    finally:
+        # Clean up temporary PDF directory
+        if 'temp_dir' in locals() and temp_dir.exists():
+            shutil.rmtree(temp_dir)
+            logger.info(f"Cleaned up temporary directory: {temp_dir}")
+        # Optionally, clean up the main project_folder if an error occurred before SB3 creation
+        # (Be careful with this if you want to inspect failed project folders)
+        # if project_folder and project_folder.exists() and sb3_file_path is None:
+        #     shutil.rmtree(project_folder)
+        #     logger.info(f"Cleaned up partial project folder: {project_folder}")
+
+@app.route('/list_projects', methods=['GET'])
+def list_projects():
+    """
+    Lists all generated .sb3 files and their download URLs for debugging.
+    """
+    try:
+        sb3_files = []
+        for file_path in GEN_PROJECT_DIR.iterdir():
+            if file_path.suffix == '.sb3' and file_path.is_file():
+                project_id = file_path.stem
+                download_url = f"/download_sb3/{project_id}"
+                sb3_files.append({"project_id": project_id, "download_url": download_url, "filename": file_path.name})
+        
+        return jsonify({"generated_sb3_files": sb3_files}), 200
+    except Exception as e:
+        logger.error(f"Error listing projects: {e}", exc_info=True)
+        return jsonify({"error": "Failed to list generated projects"}), 500
+
 if __name__ == '__main__':
-    # os.makedirs("outputs", exist_ok=True) #== commented by P
     app.run(host='0.0.0.0', port=7860, debug=True)