Update app.py

app.py

@@ -1833,7 +1833,7 @@ def similarity_matching(input_json_path: str, project_folder:str) -> str:
     #         DEFINE PATHS           #
     # ============================== #
     image_dirs = [backdrop_images_path, sprite_images_path]
-
+    print(" --------------------------------------- input_json_path ---------------------------------------", input_json_path)
     project_json_path = os.path.join(project_folder, "project.json")
     
     # ============================== #
@@ -1869,6 +1869,7 @@ def similarity_matching(input_json_path: str, project_folder:str) -> str:
     SPRITE_DIR / "Crab.sprite3" / "bear_element.png",
     SPRITE_DIR / "Soccer Ball.sprite3" / "cat_football.png",
     ]
+    print(" --------------------------------------- folder_image_paths ---------------------------------------\n", folder_image_paths)
     # ============================== #
     #      DECODE SPRITE IMAGES      #
     # ============================== #
@@ -2121,41 +2122,45 @@ def upscale_image(image: Image.Image, scale: int = 2) -> Image.Image:
         logger.error(f"❌ Error during image upscaling: {str(e)}")
         return image 
         
-def create_sb3_archive(project_folder: Path, project_id: str) -> Path | None:
+def create_sb3_archive(project_folder, project_id):
     """
     Zips the project folder and renames it to an .sb3 file.
+    
     Args:
-        project_folder (Path): The path to the directory containing the project.json and assets.
+        project_folder (str): 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:
-        Path: The path to the created .sb3 file, or None if an error occurred.
+        str: The path to the created .sb3 file, or None if an error occurred.
     """
     print(" --------------------------------------- create_sb3_archive INITIALIZE ---------------------------------------")
-    output_base_name = GEN_PROJECT_DIR / project_id
+    # output_filename = os.path.join("outputs", project_id)
+    #output_filename = OUTPUT_DIR / project_id
+    output_filename = GEN_PROJECT_DIR / project_id
     zip_path = None
     sb3_path = None
     try:
-        # shutil.make_archive automatically adds .zip extension
-        print(" --------------------------------------- zip_path_str ---------------------------------------", output_base_name, project_folder)
-        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
+        zip_path = shutil.make_archive(output_filename, 'zip', root_dir=project_folder)
+         print(" --------------------------------------- zip_path_str ---------------------------------------", output_base_name, project_folder)
         logger.info(f"Project folder zipped to: {zip_path}")
-
+        
         # 2. Rename the .zip file to .sb3
-        sb3_path = GEN_PROJECT_DIR / f"{project_id}.sb3"
+        sb3_path = f"{output_filename}.sb3"
         os.rename(zip_path, sb3_path)
+        print(" --------------------------------------- rename paths ---------------------------------------", 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}", exc_info=True)
+        logger.error(f"Error creating SB3 archive for {project_id}: {e}")
         # Clean up any partial files if an error occurs
-        if zip_path and zip_path.exists():
+        if zip_path and os.path.exists(zip_path):
             os.remove(zip_path)
-        if sb3_path and sb3_path.exists():
+        if sb3_path and os.path.exists(sb3_path):
             os.remove(sb3_path)
         return None
 
+
 @app.route('/')
 def index():
     return render_template('app_index.html')
@@ -2166,30 +2171,28 @@ def download_sb3(project_id):
     Allows users to download the generated .sb3 Scratch project file.
     """
     sb3_filename = f"{project_id}.sb3"
-    sb3_filepath = GEN_PROJECT_DIR / sb3_filename
+    sb3_filepath = os.path.join("generated_projects", sb3_filename)
 
     try:
-        if sb3_filepath.exists():
+        if os.path.exists(sb3_filepath):
             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=GEN_PROJECT_DIR,
+                directory="generated_projects",
                 path=sb3_filename,
-                as_attachment=True,
-                download_name=sb3_filename
+                as_attachment=True,  # This makes the browser download the file
+                download_name=sb3_filename # This sets the filename for the download
             )
         else:
-            logger.warning(f"SB3 file not found for ID: {project_id} at {sb3_filepath}")
+            logger.warning(f"SB3 file not found for ID: {project_id}")
             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}", exc_info=True)
+        logger.error(f"Error serving SB3 file for ID {project_id}: {e}")
         return jsonify({"error": "Failed to retrieve Scratch project file"}), 500
 
+# API endpoint
 @app.route('/process_pdf', methods=['POST'])
 def process_pdf():
-    project_id = None
-    project_folder = None
-    temp_dir = None
-    extracted_json_parent_dir = None # Initialize for finally block cleanup or later use
     try:
         logger.info("Received request to process PDF.")
         if 'pdf_file' not in request.files:
@@ -2200,48 +2203,59 @@ def process_pdf():
         if pdf_file.filename == '':
             return jsonify({"error": "Empty filename"}), 400
 
+        # ================================================= #
+        #   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.mkdir(parents=True, exist_ok=True)
-        logger.info(f"Created project folder: {project_folder}")
-
+        # =========================================================================== #
+        #           Create empty json in project_{random_id} folder                   #
+        # =========================================================================== #
+        #os.makedirs(project_folder, exist_ok=True)
+        
+        # Save the uploaded PDF temporarily
         filename = secure_filename(pdf_file.filename)
-        temp_dir = Path(tempfile.mkdtemp())
-        saved_pdf_path = temp_dir / filename
+        temp_dir = tempfile.mkdtemp()
+        saved_pdf_path = os.path.join(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}")
         
-        # MODIFIED CALL: Pass JSON_DIR and DETECTED_IMAGE_DIR
-        # extract_images_from_pdf now returns the file path to extracted_sprites.json
-        # and the parent directory where it was created.
-        extracted_sprites_json_file_path, extracted_json_parent_dir = extract_images_from_pdf(
-            saved_pdf_path, JSON_DIR, DETECTED_IMAGE_DIR
-        )
-        
-        # Now, directly use extracted_sprites_json_file_path to check for its existence
-        if not extracted_sprites_json_file_path.exists():
-            logger.error(f"No extracted_sprites.json found at {extracted_sprites_json_file_path}")
-            return jsonify({"error": "No extracted_sprites.json found"}), 500
+        # Extract & process
+        json_path = None
+        output_path, result = extract_images_from_pdf(saved_pdf_path, json_path)
+        print(" --------------------------------------- zip_path_str ---------------------------------------", output_path, result)
+        # 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")
 
-        with open(extracted_sprites_json_file_path, 'r') as f:
-            sprite_data = json.load(f)
+        # if not os.path.exists(extracted_sprites_json):
+        #     return jsonify({"error": "No extracted_sprites.json found"}), 500
 
-        # MODIFIED CALL: Pass the extracted_json_parent_dir (the directory) to similarity_matching
-        project_output = similarity_matching(extracted_json_parent_dir, project_folder)
-        logger.info("Similarity matching completed.")
+        # with open(extracted_sprites_json, 'r') as f:
+        #     sprite_data = json.load(f)
+
+        project_output = similarity_matching(output_path, project_folder)
+        logger.info("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)
+            
+        images = convert_from_path(saved_pdf_path, dpi=300)  
+        print(type)
         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",
@@ -2252,59 +2266,31 @@ def process_pdf():
         #     "temporary_node": {},
         # }
         
-        # final_state_dict = app_graph.invoke(initial_state_dict)
+        # final_state_dict = app_graph.invoke(initial_state_dict) # Pass dictionary
 
-        final_project_json = project_skeleton # final_state_dict['project_json']
+        # final_project_json = final_state_dict['project_json'] # Access as dict
+        final_project_json = project_skeleton
         
+        # Save the *final* filled project JSON, overwriting the skeleton
         with open(project_output, "w") as f:
             json.dump(final_project_json, f, indent=2)
         logger.info(f"Final project JSON saved to {project_output}")
 
+        # --- Call the new function to create the .sb3 file ---
         sb3_file_path = create_sb3_archive(project_folder, project_id)
         if sb3_file_path:
             logger.info(f"Successfully created SB3 file: {sb3_file_path}")
-            download_url = f"/download_sb3/{project_id}"
+            # 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}" 
             print(f"DOWNLOAD_URL: {download_url}")
-            return jsonify({"message": "Processed PDF and Game sb3 generated successfully", "project_id": project_id, "download_url": download_url})
+            return jsonify({"message": "Procesed 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
-    finally:
-        if temp_dir and temp_dir.exists():
-            shutil.rmtree(temp_dir)
-            logger.info(f"Cleaned up temporary directory: {temp_dir}")
-        
-        # Optional: Clean up the extracted JSON and image directories for this project_id
-        # if extracted_json_parent_dir and extracted_json_parent_dir.exists():
-        #     shutil.rmtree(extracted_json_parent_dir)
-        #     logger.info(f"Cleaned up extracted JSON directory: {extracted_json_parent_dir}")
-        # if pdf_filename_stem: # You'd need to get pdf_filename_stem from `filename` earlier
-        #     corresponding_image_dir = DETECTED_IMAGE_DIR / pdf_filename_stem
-        #     if corresponding_image_dir.exists():
-        #         shutil.rmtree(corresponding_image_dir)
-        #         logger.info(f"Cleaned up detected image directory: {corresponding_image_dir}")
-
-
-@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
-
+        return jsonify({"error": f"❌ Failed to process PDF: {str(e)}"}), 500   
+    
 if __name__ == '__main__':
+    # os.makedirs("outputs", exist_ok=True) #== commented by P
     app.run(host='0.0.0.0', port=7860, debug=True)