VisionFlowAutomator/main.py

import sys
import os
import json
import threading
import time
import logging
import zipfile
import shutil
import gc
import weakref
from functools import lru_cache
from PyQt6 import QtWidgets, QtGui, QtCore
import cv2
import pygetwindow as gw
import numpy as np
import pyautogui
from pynput import keyboard
import tkinter as tk
from PIL import ImageGrab, Image, ImageTk

# Disable the PyAutoGUI fail-safe feature and optimize settings
pyautogui.FAILSAFE = False
pyautogui.PAUSE = 0.01  # Reduce pause between actions
pyautogui.MINIMUM_DURATION = 0  # Remove minimum duration for faster execution

# Setup logs directory
LOGS_DIR = 'logs'
if not os.path.exists(LOGS_DIR):
    os.makedirs(LOGS_DIR)

# Setup logging with optimized settings
logging.basicConfig(
    level=logging.INFO,  # Changed from DEBUG to reduce log volume
    format='[%(asctime)s] %(levelname)s: %(message)s',
    handlers=[
        logging.FileHandler(os.path.join(LOGS_DIR, 'scenario_automation.log'), encoding='utf-8'),
        logging.StreamHandler()
    ]
)

# Create a more efficient logger for performance-critical sections
perf_logger = logging.getLogger('performance')
perf_logger.setLevel(logging.WARNING)  # Only log warnings and errors for performance

logger = logging.getLogger(__name__)

def get_gpu_memory_info():
    """
    Get GPU memory information using multiple methods.
    Supports both discrete and integrated GPUs across Windows, Linux, and macOS.
    Returns dict with GPU memory info or basic info if no GPU found.
    """
    gpu_info = {'has_gpu': False, 'total_mb': 0, 'used_mb': 0, 'free_mb': 0, 'utilization_percent': 0, 'gpu_name': 'Unknown'}
    
    # Try nvidia-ml-py (NVIDIA GPUs - most detailed info)
    try:
        import pynvml
        pynvml.nvmlInit()
        handle = pynvml.nvmlDeviceGetHandleByIndex(0)  # Get first GPU
        
        # Get memory info
        mem_info = pynvml.nvmlDeviceGetMemoryInfo(handle)
        total_mb = mem_info.total / 1024 / 1024
        used_mb = mem_info.used / 1024 / 1024
        free_mb = mem_info.free / 1024 / 1024
        utilization_percent = (used_mb / total_mb) * 100
        
        # Get GPU name
        gpu_name = pynvml.nvmlDeviceGetName(handle).decode('utf-8')
        
        gpu_info.update({
            'has_gpu': True,
            'total_mb': total_mb,
            'used_mb': used_mb,
            'free_mb': free_mb,
            'utilization_percent': utilization_percent,
            'gpu_name': gpu_name,
            'method': 'pynvml'
        })
        logger.debug(f"GPU detected via pynvml: {gpu_name}, {total_mb:.0f}MB total")
        return gpu_info
        
    except (ImportError, Exception) as e:
        logger.debug(f"pynvml not available or failed: {e}")
    
    # Try GPUtil (NVIDIA GPUs alternative)
    try:
        import GPUtil
        gpus = GPUtil.getGPUs()
        if gpus:
            gpu = gpus[0]  # Get first GPU
            total_mb = gpu.memoryTotal
            used_mb = gpu.memoryUsed
            free_mb = gpu.memoryFree
            utilization_percent = (used_mb / total_mb) * 100
            
            gpu_info.update({
                'has_gpu': True,
                'total_mb': total_mb,
                'used_mb': used_mb,
                'free_mb': free_mb,
                'utilization_percent': utilization_percent,
                'gpu_name': gpu.name,
                'method': 'GPUtil'
            })
            logger.debug(f"GPU detected via GPUtil: {gpu.name}, {total_mb:.0f}MB total")
            return gpu_info
            
    except (ImportError, Exception) as e:
        logger.debug(f"GPUtil not available or failed: {e}")
    
    # Try psutil for basic GPU info (limited)
    try:
        import psutil
        # Check if there are any GPU-related processes
        for proc in psutil.process_iter(['pid', 'name']):
            try:
                proc_name = proc.info['name'].lower()
                if any(gpu_process in proc_name for gpu_process in ['nvidia', 'amd', 'intel', 'gpu']):
                    gpu_info.update({
                        'has_gpu': True,
                        'total_mb': 0,  # Can't get detailed info with psutil
                        'used_mb': 0,
                        'free_mb': 0,
                        'utilization_percent': 0,
                        'gpu_name': 'Detected via process',
                        'method': 'psutil_detection'
                    })
                    logger.debug(f"GPU detected via psutil process detection: {proc_name}")
                    return gpu_info
            except (psutil.NoSuchProcess, psutil.AccessDenied):
                continue
                
    except ImportError:
        pass
    
    # Try Windows-specific methods
    try:
        import subprocess
        import re
        
        # Try nvidia-smi command for NVIDIA GPUs
        try:
            result = subprocess.run(['nvidia-smi', '--query-gpu=name,memory.total,memory.used,memory.free', '--format=csv,noheader,nounits'], 
                                  capture_output=True, text=True, timeout=5)
            
            if result.returncode == 0 and result.stdout.strip():
                lines = result.stdout.strip().split('\n')
                if lines:
                    parts = lines[0].split(', ')
                    if len(parts) >= 4:
                        gpu_name = parts[0].strip()
                        total_mb = float(parts[1].strip())
                        used_mb = float(parts[2].strip())
                        free_mb = float(parts[3].strip())
                        utilization_percent = (used_mb / total_mb) * 100
                        
                        gpu_info.update({
                            'has_gpu': True,
                            'total_mb': total_mb,
                            'used_mb': used_mb,
                            'free_mb': free_mb,
                            'utilization_percent': utilization_percent,
                            'gpu_name': gpu_name,
                            'method': 'nvidia-smi'
                        })
                        return gpu_info
                        
        except (subprocess.TimeoutExpired, subprocess.CalledProcessError, FileNotFoundError):
            pass
        
        # Try Intel GPU detection via Windows Management Instrumentation (WMI)
        try:
            # Query for Intel integrated graphics
            wmi_result = subprocess.run([
                'powershell', '-Command',
                "Get-WmiObject -Class Win32_VideoController | Where-Object {$_.Name -like '*Intel*' -or $_.Name -like '*UHD*' -or $_.Name -like '*Iris*' -or $_.Name -like '*HD Graphics*'} | Select-Object Name, AdapterRAM | ConvertTo-Json"
            ], capture_output=True, text=True, timeout=10)
            
            if wmi_result.returncode == 0 and wmi_result.stdout.strip():
                import json
                wmi_data = json.loads(wmi_result.stdout.strip())
                
                # Handle both single GPU and multiple GPUs
                if isinstance(wmi_data, dict):
                    wmi_data = [wmi_data]
                
                for gpu_data in wmi_data:
                    if gpu_data.get('Name') and gpu_data.get('AdapterRAM'):
                        gpu_name = gpu_data['Name']
                        # AdapterRAM is in bytes, convert to MB
                        total_mb = int(gpu_data['AdapterRAM']) / (1024 * 1024)
                        
                        gpu_info.update({
                            'has_gpu': True,
                            'total_mb': total_mb,
                            'used_mb': 0,  # WMI doesn't provide usage info
                            'free_mb': total_mb,  # Assume all free since we can't get usage
                            'utilization_percent': 0,
                            'gpu_name': gpu_name,
                            'method': 'WMI_Intel'
                        })
                        logger.debug(f"Intel GPU detected via WMI: {gpu_name}, {total_mb:.0f}MB")
                        return gpu_info
                        
        except (subprocess.TimeoutExpired, subprocess.CalledProcessError, FileNotFoundError, json.JSONDecodeError):
            pass
        
        # Try AMD GPU detection via WMI
        try:
            wmi_result = subprocess.run([
                'powershell', '-Command',
                "Get-WmiObject -Class Win32_VideoController | Where-Object {$_.Name -like '*AMD*' -or $_.Name -like '*Radeon*' -or $_.Name -like '*ATI*'} | Select-Object Name, AdapterRAM | ConvertTo-Json"
            ], capture_output=True, text=True, timeout=10)
            
            if wmi_result.returncode == 0 and wmi_result.stdout.strip():
                import json
                wmi_data = json.loads(wmi_result.stdout.strip())
                
                # Handle both single GPU and multiple GPUs
                if isinstance(wmi_data, dict):
                    wmi_data = [wmi_data]
                
                for gpu_data in wmi_data:
                    if gpu_data.get('Name') and gpu_data.get('AdapterRAM'):
                        gpu_name = gpu_data['Name']
                        # AdapterRAM is in bytes, convert to MB
                        total_mb = int(gpu_data['AdapterRAM']) / (1024 * 1024)
                        
                        gpu_info.update({
                            'has_gpu': True,
                            'total_mb': total_mb,
                            'used_mb': 0,  # WMI doesn't provide usage info
                            'free_mb': total_mb,  # Assume all free since we can't get usage
                            'utilization_percent': 0,
                            'gpu_name': gpu_name,
                            'method': 'WMI_AMD'
                        })
                        return gpu_info
                        
        except (subprocess.TimeoutExpired, subprocess.CalledProcessError, FileNotFoundError, json.JSONDecodeError):
            pass
        
        # Generic WMI query for any video controller
        try:
            wmi_result = subprocess.run([
                'powershell', '-Command',
                "Get-WmiObject -Class Win32_VideoController | Where-Object {$_.AdapterRAM -gt 0} | Select-Object Name, AdapterRAM | ConvertTo-Json"
            ], capture_output=True, text=True, timeout=10)
            
            if wmi_result.returncode == 0 and wmi_result.stdout.strip():
                import json
                wmi_data = json.loads(wmi_result.stdout.strip())
                
                # Handle both single GPU and multiple GPUs
                if isinstance(wmi_data, dict):
                    wmi_data = [wmi_data]
                
                # Find the first GPU with significant memory (>= 512MB)
                for gpu_data in wmi_data:
                    if gpu_data.get('Name') and gpu_data.get('AdapterRAM'):
                        total_mb = int(gpu_data['AdapterRAM']) / (1024 * 1024)
                        
                        # Only consider GPUs with at least 512MB
                        if total_mb >= 512:
                            gpu_name = gpu_data['Name']
                            
                            gpu_info.update({
                                'has_gpu': True,
                                'total_mb': total_mb,
                                'used_mb': 0,  # WMI doesn't provide usage info
                                'free_mb': total_mb,  # Assume all free since we can't get usage
                                'utilization_percent': 0,
                                'gpu_name': gpu_name,
                                'method': 'WMI_Generic'
                            })
                            return gpu_info
                        
        except (subprocess.TimeoutExpired, subprocess.CalledProcessError, FileNotFoundError, json.JSONDecodeError):
            pass
            
    except Exception as e:
        logger.debug(f"Windows GPU detection methods failed: {e}")
    
    # Try DirectX DXGI detection (Windows 10+)
    try:
        import subprocess
        
        # Use PowerShell to query DirectX information
        dxdiag_result = subprocess.run([
            'powershell', '-Command',
            """
            Add-Type -AssemblyName System.Windows.Forms
            $dxdiag = Get-WmiObject -Class Win32_VideoController | Where-Object {$_.AdapterRAM -gt 536870912} | Select-Object Name, AdapterRAM, DriverVersion
            $dxdiag | ConvertTo-Json
            """
        ], capture_output=True, text=True, timeout=15)
        
        if dxdiag_result.returncode == 0 and dxdiag_result.stdout.strip():
            import json
            dx_data = json.loads(dxdiag_result.stdout.strip())
            
            if isinstance(dx_data, dict):
                dx_data = [dx_data]
            
            for gpu_data in dx_data:
                if gpu_data.get('Name') and gpu_data.get('AdapterRAM'):
                    gpu_name = gpu_data['Name']
                    total_mb = int(gpu_data['AdapterRAM']) / (1024 * 1024)
                    
                    gpu_info.update({
                        'has_gpu': True,
                        'total_mb': total_mb,
                        'used_mb': 0,
                        'free_mb': total_mb,
                        'utilization_percent': 0,
                        'gpu_name': gpu_name,
                        'method': 'DirectX_DXGI'
                    })
                    return gpu_info
                    
    except (subprocess.TimeoutExpired, subprocess.CalledProcessError, FileNotFoundError, json.JSONDecodeError):
        pass
    
    # Try Linux/macOS methods for integrated GPUs
    try:
        import subprocess
        import platform
        
        system = platform.system().lower()
        
        if system == 'linux':
            # Try lspci for GPU detection on Linux
            try:
                lspci_result = subprocess.run(['lspci', '-v'], capture_output=True, text=True, timeout=10)
                if lspci_result.returncode == 0:
                    output = lspci_result.stdout.lower()
                    
                    # Look for VGA/Display controllers
                    if any(keyword in output for keyword in ['vga', 'display', 'gpu', 'graphics']):
                        # Extract GPU names
                        lines = lspci_result.stdout.split('\n')
                        for line in lines:
                            if any(keyword in line.lower() for keyword in ['vga', 'display controller', '3d controller']):
                                # Try to extract GPU name from the line
                                if 'intel' in line.lower():
                                    gpu_name = 'Intel Integrated Graphics (detected via lspci)'
                                elif 'amd' in line.lower() or 'ati' in line.lower():
                                    gpu_name = 'AMD Integrated Graphics (detected via lspci)'
                                elif 'nvidia' in line.lower():
                                    gpu_name = 'NVIDIA GPU (detected via lspci)'
                                else:
                                    gpu_name = 'GPU detected via lspci'
                                
                                gpu_info.update({
                                    'has_gpu': True,
                                    'total_mb': 0,  # Can't get memory info from lspci
                                    'used_mb': 0,
                                    'free_mb': 0,
                                    'utilization_percent': 0,
                                    'gpu_name': gpu_name,
                                    'method': 'lspci_linux'
                                })
                                return gpu_info
                                
            except (subprocess.TimeoutExpired, subprocess.CalledProcessError, FileNotFoundError):
                pass
            
            # Try checking /sys/class/drm for Intel iGPU on Linux
            try:
                import os
                drm_path = '/sys/class/drm'
                if os.path.exists(drm_path):
                    drm_devices = os.listdir(drm_path)
                    intel_devices = [d for d in drm_devices if 'i915' in d or 'intel' in d.lower()]
                    
                    if intel_devices:
                        gpu_info.update({
                            'has_gpu': True,
                            'total_mb': 0,
                            'used_mb': 0,
                            'free_mb': 0,
                            'utilization_percent': 0,
                            'gpu_name': 'Intel Integrated Graphics (detected via /sys/class/drm)',
                            'method': 'linux_drm'
                        })
                        return gpu_info
                        
            except Exception:
                pass
        
        elif system == 'darwin':  # macOS
            # Try system_profiler for macOS GPU detection
            try:
                profiler_result = subprocess.run([
                    'system_profiler', 'SPDisplaysDataType', '-json'
                ], capture_output=True, text=True, timeout=15)
                
                if profiler_result.returncode == 0:
                    import json
                    display_data = json.loads(profiler_result.stdout)
                    
                    displays = display_data.get('SPDisplaysDataType', [])
                    for display in displays:
                        gpu_name = display.get('sppci_model', 'Unknown GPU')
                        vram = display.get('spdisplays_vram', '0 MB')
                        
                        # Extract VRAM size
                        vram_mb = 0
                        if 'MB' in vram:
                            try:
                                vram_mb = float(vram.replace(' MB', ''))
                            except:
                                pass
                        elif 'GB' in vram:
                            try:
                                vram_mb = float(vram.replace(' GB', '')) * 1024
                            except:
                                pass
                        
                        gpu_info.update({
                            'has_gpu': True,
                            'total_mb': vram_mb,
                            'used_mb': 0,
                            'free_mb': vram_mb,
                            'utilization_percent': 0,
                            'gpu_name': gpu_name,
                            'method': 'macos_system_profiler'
                        })
                        return gpu_info
                        
            except (subprocess.TimeoutExpired, subprocess.CalledProcessError, FileNotFoundError, json.JSONDecodeError):
                pass
                
    except Exception as e:
        logger.debug(f"Linux/macOS GPU detection failed: {e}")
    
    return gpu_info

CONFIG_DIR = 'scenarios'
if not os.path.exists(CONFIG_DIR):
    os.makedirs(CONFIG_DIR)

class TemplateCache:
    """
    Cache for template images to avoid reloading them repeatedly.
    Uses weak references to allow garbage collection when templates are no longer needed.
    """
    def __init__(self, max_size=50):
        self._cache = {}
        self._max_size = max_size
        self._access_times = {}
    
    def get_template(self, path, force_reload=False):
        """Get template image from cache or load it."""
        if force_reload or path not in self._cache:
            if len(self._cache) >= self._max_size:
                self._cleanup_old_entries()
            
            template = cv2.imread(path)
            if template is not None:
                self._cache[path] = template
                self._access_times[path] = time.time()
            return template
        else:
            self._access_times[path] = time.time()
            return self._cache[path]
    
    def _cleanup_old_entries(self):
        """Remove least recently used entries."""
        if not self._access_times:
            return
        
        # Remove oldest 25% of entries
        sorted_entries = sorted(self._access_times.items(), key=lambda x: x[1])
        entries_to_remove = len(sorted_entries) // 4
        
        for path, _ in sorted_entries[:entries_to_remove]:
            self._cache.pop(path, None)
            self._access_times.pop(path, None)
    
    def clear(self):
        """Clear all cached templates."""
        self._cache.clear()
        self._access_times.clear()
        gc.collect()

# Global template cache instance
template_cache = TemplateCache()

class Scenario:
    """
    Represents a scenario, which consists of a name and a list of steps.
    Handles saving/loading scenario data to/from disk.
    """
    def __init__(self, name, steps=None):
        """
        Initialize a Scenario with a name and optional steps.
        """
        self.name = name
        self.steps = steps or []  # List of Step dicts

    def to_dict(self):
        """
        Return a dictionary representation of the scenario.
        """
        return {'name': self.name, 'steps': self.steps}

    @staticmethod
    def from_dict(data):
        """
        Create a Scenario object from a dictionary.
        """
        return Scenario(data['name'], data.get('steps', []))

    def get_scenario_dir(self):
        """
        Return the directory path for this scenario's files.
        """
        return os.path.join(CONFIG_DIR, self.name)

    def save(self):
        """
        Save the scenario to disk as a JSON file.
        """
        try:
            scenario_dir = self.get_scenario_dir()
            if not os.path.exists(scenario_dir):
                os.makedirs(scenario_dir)
            
            with open(os.path.join(scenario_dir, 'scenario.json'), 'w') as f:
                json.dump(self.to_dict(), f, indent=2)
            logger.info(f"Scenario '{self.name}' saved.")
        except Exception as e:
            logger.error(f"Failed to save scenario '{self.name}': {e}")

    @staticmethod
    def load(name):
        """
        Load a scenario from disk by name.
        """
        try:
            scenario_dir = os.path.join(CONFIG_DIR, name)
            with open(os.path.join(scenario_dir, 'scenario.json'), 'r') as f:
                scenario = Scenario.from_dict(json.load(f))
            logger.info(f"Scenario '{name}' loaded.")
            return scenario
        except Exception as e:
            logger.error(f"Failed to load scenario '{name}': {e}")
            return None

    @staticmethod
    def list_all():
        """
        List all scenario directories.
        """
        return [d for d in os.listdir(CONFIG_DIR) if os.path.isdir(os.path.join(CONFIG_DIR, d))]

def take_screenshot_with_tkinter():
    """
    Use Tkinter to let the user select a region of the screen for a screenshot.
    Optimized for better performance and memory usage.
    Returns a dict with x, y, width, height, or None.
    """
    root = tk.Tk()
    root.attributes("-alpha", 0.3)
    root.attributes("-fullscreen", True)
    root.attributes("-topmost", True)  # Ensure window stays on top
    root.wait_visibility(root)
    
    # Configure for better performance
    root.resizable(False, False)
    root.overrideredirect(True)
    
    canvas = tk.Canvas(root, cursor="cross", highlightthickness=0)
    canvas.pack(fill="both", expand=True)

    rect = None
    start_x = None
    start_y = None
    selection_rect = None
    
    # Add instructions
    instruction_text = canvas.create_text(
        root.winfo_screenwidth() // 2, 50, 
        text="Click and drag to select area. Press ESC to cancel.", 
        fill="white", font=("Arial", 14)
    )

    def on_button_press(event):
        nonlocal start_x, start_y, rect
        start_x = event.x
        start_y = event.y
        # Remove instruction text
        canvas.delete(instruction_text)
        rect = canvas.create_rectangle(start_x, start_y, start_x, start_y, outline='red', width=2)

    def on_mouse_drag(event):
        nonlocal rect
        if rect and start_x is not None and start_y is not None:
            cur_x, cur_y = event.x, event.y
            canvas.coords(rect, start_x, start_y, cur_x, cur_y)

    def on_button_release(event):
        nonlocal selection_rect
        if start_x is not None and start_y is not None:
            end_x, end_y = event.x, event.y
            
            x1 = min(start_x, end_x)
            y1 = min(start_y, end_y)
            x2 = max(start_x, end_x)
            y2 = max(start_y, end_y)

            width = x2 - x1
            height = y2 - y1

            # Minimum size validation
            if width > 10 and height > 10:
                selection_rect = {"x": x1, "y": y1, "width": width, "height": height}
            else:
                selection_rect = None
        
        root.quit()

    def on_escape(event):
        nonlocal selection_rect
        selection_rect = None
        root.quit()
    
    def on_key_press(event):
        if event.keysym == 'Escape':
            on_escape(event)

    # Bind events
    canvas.bind("<ButtonPress-1>", on_button_press)
    canvas.bind("<B1-Motion>", on_mouse_drag)
    canvas.bind("<ButtonRelease-1>", on_button_release)
    root.bind("<Escape>", on_escape)
    root.bind("<KeyPress>", on_key_press)
    
    # Focus the window to receive key events
    root.focus_set()
    root.focus_force()

    try:
        root.mainloop()
    except Exception as e:
        logger.error(f"Error in screenshot selection: {e}")
        selection_rect = None
    finally:
        try:
            root.destroy()
        except Exception as e:
            logger.warning(f"Error destroying tkinter window: {e}")
    
    return selection_rect

class MainWindow(QtWidgets.QMainWindow):
    """
    Main application window for Scenario Image Automation.
    Handles UI, scenario management, and automation logic.
    """
    def automation_loop(self):
        """
        Optimized automation loop with memory management and performance improvements.
        """
        logger.info('Automation loop started.')
        frame_count = 0
        last_gc_time = time.time()
        
        try:
            while self.running:
                loop_start_time = time.time()
                
                # Only update the state label periodically to reduce UI overhead
                if frame_count % 10 == 0:
                    self.set_state('Looking for matches')
                
                # Get optimized screenshot
                screen_data = self._get_optimized_screenshot()
                if screen_data is None:
                    time.sleep(0.1)
                    continue
                
                screen_np, offset_x, offset_y = screen_data
                
                # Process steps with priority and cooldown system
                step_executed = False
                current_time = time.time()
                
                for step_idx, step in enumerate(self.current_scenario.steps):
                    # Check step cooldown to prevent spam
                    step_name = step.get('name', f'step_{step_idx}')
                    if step_name in self._step_cooldown:
                        if current_time - self._step_cooldown[step_name] < 1.0:
                            continue
                    
                    # Process step images with cache
                    detections = self._process_step_images(step, screen_np, offset_x, offset_y)
                    
                    # Check trigger condition
                    if self._check_step_trigger(step, detections):
                        self._processing_step = True
                        self.set_state(f'Performing step: {step_name}')
                        logger.info(f"Executing step: {step_name}")
                        
                        # Execute step actions
                        success = self._execute_step_actions(step, detections)
                        
                        if success:
                            self._step_cooldown[step_name] = current_time
                            step_executed = True
                            
                        self._processing_step = False
                        break  # Execute only one step per loop iteration
                
                # Performance monitoring
                loop_time = time.time() - loop_start_time
                self._update_performance_stats(loop_time)
                
                # Periodic cleanup
                frame_count += 1
                if frame_count % 100 == 0:  # Every 100 frames
                    self._periodic_cleanup()
                
                # Dynamic sleep based on performance
                sleep_time = max(0.05, 0.2 - loop_time) if not step_executed else 0.1
                time.sleep(sleep_time)
                
        except Exception as e:
            logger.error(f'Automation loop error: {e}')
        finally:
            self._processing_step = False
            self.set_state('Paused')
            logger.info('Automation loop ended.')
    
    def _get_optimized_screenshot(self):
        """
        Get screenshot with caching to reduce memory allocation.
        """
        current_time = time.time()
        
        # Use cached screenshot if recent enough and not processing
        if (self._last_screenshot is not None and 
            not self._processing_step and
            current_time - self._last_screenshot_time < self._screenshot_cache_duration):
            return self._last_screenshot
        
        try:
            selected_window = self.window_combo.currentText()
            if selected_window == 'Entire Screen':
                screen = pyautogui.screenshot()
                offset_x, offset_y = 0, 0
            else:
                # Try to get specific window
                win = None
                for w in gw.getAllWindows():
                    if w.title == selected_window:
                        win = w
                        break
                
                if win is not None and win.width > 0 and win.height > 0:
                    x, y, w_, h_ = win.left, win.top, win.width, win.height
                    # Validate window bounds
                    if x >= 0 and y >= 0 and w_ > 0 and h_ > 0:
                        screen = pyautogui.screenshot(region=(x, y, w_, h_))
                        offset_x, offset_y = x, y
                    else:
                        screen = pyautogui.screenshot()
                        offset_x, offset_y = 0, 0
                else:
                    screen = pyautogui.screenshot()
                    offset_x, offset_y = 0, 0
            
            # Convert to OpenCV format with optimized memory usage
            screen_np = cv2.cvtColor(np.array(screen), cv2.COLOR_RGB2BGR)
            
            # Cache the result
            screen_data = (screen_np, offset_x, offset_y)
            self._last_screenshot = screen_data
            self._last_screenshot_time = current_time
            
            # Clean up PIL image to free memory
            del screen
            
            return screen_data
            
        except Exception as e:
            logger.error(f'Error capturing screenshot: {e}')
            return None
    
    def _process_step_images(self, step, screen_np, offset_x, offset_y):
        """
        Process step images with template caching.
        """
        detections = {}
        
        for img in step.get('images', []):
            try:
                # Use cached template
                template = template_cache.get_template(img['path'])
                if template is None:
                    logger.warning(f"Could not load template image: {img['path']}")
                    continue
                
                # Perform template matching with optimization
                res = cv2.matchTemplate(screen_np, template, cv2.TM_CCOEFF_NORMED)
                min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
                
                sensitivity = img.get('sensitivity', 0.9)
                if max_val > sensitivity:
                    # Adjust detected location to be relative to the full screen
                    abs_loc = (max_loc[0] + offset_x, max_loc[1] + offset_y)
                    detections[img['name']] = (abs_loc, template.shape, max_val)
                
                # Clean up OpenCV result to free memory
                del res
                
            except Exception as e:
                logger.error(f"Error in image detection for {img.get('name', 'unknown')}: {e}")
        
        return detections
    
    def _check_step_trigger(self, step, detections):
        """
        Check if step should be triggered based on detections.
        """
        step_images = step.get('images', [])
        if not step_images:
            return False
        
        found = [img for img in step_images if img.get('name') in detections]
        condition = step.get('condition', 'OR')
        
        if condition == 'AND':
            return len(found) == len(step_images)
        else:  # OR condition
            return len(found) > 0
    
    def _execute_step_actions(self, step, detections):
        """
        Execute step actions with error handling.
        """
        try:
            found_images = [img for img in step.get('images', []) if img.get('name') in detections]
            if found_images:
                # Use the first detected image for position reference
                ref_img = found_images[0]
                loc, shape, confidence = detections.get(ref_img.get('name'), ((0, 0), (0, 0, 0), 0))
                
                for action in step.get('actions', []):
                    self._perform_step_action(action, loc, shape)
                
                return True
            
        except Exception as e:
            logger.error(f"Error executing step actions: {e}")
        
        return False
    
    def _update_performance_stats(self, loop_time):
        """
        Update performance statistics for monitoring.
        """
        self._loop_times.append(loop_time)
        if len(self._loop_times) > self._max_loop_time_samples:
            self._loop_times.pop(0)
        
        # Performance warnings are now handled by _monitor_performance with deduplication
    
    def _periodic_cleanup(self):
        """
        Perform periodic cleanup to prevent memory leaks.
        """
        # Clear old screenshot cache
        self._last_screenshot = None
        
        # Clean up old step cooldowns (older than 5 minutes)
        current_time = time.time()
        old_cooldowns = [name for name, timestamp in self._step_cooldown.items() 
                        if current_time - timestamp > 300]
        for name in old_cooldowns:
            del self._step_cooldown[name]
        
        # Force garbage collection
        gc.collect()

    def _perform_step_action(self, action, loc, shape):
        """
        Perform a single step action (click, key, scroll, delay) at the given location.
        """
        act_type = action['type']
        params = action['params']
        logger.debug(f'Performing step action: {act_type}, params={params}, loc={loc}, shape={shape}')
        try:
            if act_type == 'click':
                pos_type = params.get('pos_type', 'center')
                if pos_type == 'center':
                    x = loc[0] + shape[1] // 2
                    y = loc[1] + shape[0] // 2
                elif pos_type == 'relative':
                    x = loc[0] + params.get('rel_x', 0)
                    y = loc[1] + params.get('rel_y', 0)
                elif pos_type == 'absolute':
                    x = params.get('abs_x', 0)
                    y = params.get('abs_y', 0)
                else:
                    x, y = loc[0], loc[1]
                button = params.get('button', 'left')
                logger.info(f'Clicking at ({x}, {y}) with button {button}')
                pyautogui.click(x, y, button=button)
            elif act_type == 'key':
                key = params.get('key', 'enter')
                logger.info(f'Pressing key: {key}')
                pyautogui.press(key)
            elif act_type == 'scroll':
                amount = params.get('amount', 0)
                direction = params.get('direction', 'up')
                logger.info(f'Scrolling: {amount} direction: {direction}')
                if direction == 'up':
                    pyautogui.scroll(amount)
                else:
                    pyautogui.scroll(-abs(amount))
            elif act_type == 'delay':
                duration = params.get('duration', 1)
                logger.info(f'Delay for {duration} seconds')
                time.sleep(duration)
        except Exception as e:
            logger.error(f'Error performing step action {act_type}: {e}')
    def set_state(self, text):
        """
        Update the state label in the UI.
        """
        self.state_text = text
        self.state_label.setText(f'State: {text}')

    def toggle_automation(self):
        """
        Toggle the automation state (start/stop). A debounce delay is added only
        when changing from stop to start, not from start to stop.
        """
        if self.running:
            self.stop_automation()
        else:
            now = time.time()
            if now - self.last_toggle_time < 5:
                logger.info("Start toggle ignored: pressed too quickly after stopping.")
                QtWidgets.QMessageBox.information(self, "Wait", "Please wait at least 5 seconds before starting again.")
                return
            self.start_automation()

    def start_automation(self):
        """
        Start the automation process and worker thread with resource checks.
        """
        logger.info('Starting automation.')
        if not self.current_scenario or self.running:
            logger.warning('Start Automation: No scenario selected or already running.')
            return
        
        # Perform pre-start cleanup
        self._last_screenshot = None
        self._step_cooldown.clear()
        gc.collect()
        
        # Initialize performance monitoring
        self._loop_times.clear()
        
        # Start automation
        self.running = True
        self.set_state('Looking for matches')
        self.btn_start_stop.setText(f'Stop ({self.hotkey.upper()})')
        
        # Create and start worker thread
        self.worker = threading.Thread(target=self.automation_loop, daemon=True)
        self.worker.start()
        
        # Setup hotkey listener with error handling
        try:
            self.listener = keyboard.GlobalHotKeys({self.hotkey: self.stop_automation})
            self.listener.start()
        except Exception as e:
            logger.error(f"Failed to setup hotkey listener: {e}")
            # Continue without hotkey if setup fails
        
        logger.info('Automation started successfully.')
    
    def get_memory_usage(self):
        """
        Get current memory usage information for monitoring, including GPU memory.
        """
        try:
            import psutil
            process = psutil.Process()
            memory_info = process.memory_info()
            cpu_percent = process.cpu_percent()
            
            # Get system info
            system_memory = psutil.virtual_memory()
            
            # Get GPU information
            gpu_info = get_gpu_memory_info()
            
            return {
                'process_memory_mb': memory_info.rss / 1024 / 1024,  # MB
                'process_memory_percent': process.memory_percent(),
                'cpu_percent': cpu_percent,
                'system_memory_percent': system_memory.percent,
                'system_memory_available_gb': system_memory.available / 1024 / 1024 / 1024,  # GB
                'template_cache_size': len(template_cache._cache) if hasattr(template_cache, '_cache') else 0,
                'cooldown_entries': len(self._step_cooldown) if hasattr(self, '_step_cooldown') else 0,
                'has_psutil': True,
                # GPU information
                'gpu_has_gpu': gpu_info['has_gpu'],
                'gpu_total_mb': gpu_info['total_mb'],
                'gpu_used_mb': gpu_info['used_mb'],
                'gpu_free_mb': gpu_info['free_mb'],
                'gpu_utilization_percent': gpu_info['utilization_percent'],
                'gpu_name': gpu_info['gpu_name'],
                'gpu_method': gpu_info.get('method', 'none')
            }
        except ImportError:
            # psutil not available, get GPU info anyway
            gpu_info = get_gpu_memory_info()
            return {
                'process_memory_mb': 0,
                'process_memory_percent': 0,
                'cpu_percent': 0,
                'system_memory_percent': 0,
                'system_memory_available_gb': 0,
                'template_cache_size': len(template_cache._cache) if hasattr(template_cache, '_cache') else 0,
                'cooldown_entries': len(self._step_cooldown) if hasattr(self, '_step_cooldown') else 0,
                'has_psutil': False,
                # GPU information
                'gpu_has_gpu': gpu_info['has_gpu'],
                'gpu_total_mb': gpu_info['total_mb'],
                'gpu_used_mb': gpu_info['used_mb'],
                'gpu_free_mb': gpu_info['free_mb'],
                'gpu_utilization_percent': gpu_info['utilization_percent'],
                'gpu_name': gpu_info['gpu_name'],
                'gpu_method': gpu_info.get('method', 'none')
            }
        except Exception as e:
            logger.warning(f"Error getting memory usage: {e}")
            # Try to get GPU info even if psutil fails
            gpu_info = get_gpu_memory_info()
            return {
                'process_memory_mb': 0,
                'process_memory_percent': 0,
                'cpu_percent': 0,
                'system_memory_percent': 0,
                'system_memory_available_gb': 0,
                'template_cache_size': len(template_cache._cache) if hasattr(template_cache, '_cache') else 0,
                'cooldown_entries': len(self._step_cooldown) if hasattr(self, '_step_cooldown') else 0,
                'has_psutil': False,
                'error': str(e),
                # GPU information
                'gpu_has_gpu': gpu_info['has_gpu'],
                'gpu_total_mb': gpu_info['total_mb'],
                'gpu_used_mb': gpu_info['used_mb'],
                'gpu_free_mb': gpu_info['free_mb'],
                'gpu_utilization_percent': gpu_info['utilization_percent'],
                'gpu_name': gpu_info['gpu_name'],
                'gpu_method': gpu_info.get('method', 'none')
            }

    def stop_automation(self):
        """
        Stop the automation process and worker thread with proper cleanup.
        """
        logger.info('Stopping automation.')
        self.running = False
        self.last_toggle_time = time.time()
        self.set_state('Paused')
        self.btn_start_stop.setText(f'Start ({self.hotkey.upper()})')
        
        # Stop global hotkey listener
        if self.listener:
            try:
                self.listener.stop()
            except Exception as e:
                logger.warning(f"Error stopping hotkey listener: {e}")
            finally:
                self.listener = None
        
        # Wait for worker thread to finish (with timeout)
        if self.worker and self.worker.is_alive():
            try:
                self.worker.join(timeout=2.0)  # Wait up to 2 seconds
                if self.worker.is_alive():
                    logger.warning("Worker thread did not stop within timeout")
            except Exception as e:
                logger.warning(f"Error joining worker thread: {e}")
            finally:
                self.worker = None
        
        # Clear processing flags and cached data
        self._processing_step = False
        self._last_screenshot = None
        self._step_cooldown.clear()
        
        # Force garbage collection to free memory
        gc.collect()
        
        logger.info('Automation stopped and resources cleaned up.')

    def __init__(self):
        """
        Initialize the main window, UI, and load scenarios.
        """
        super().__init__()
        self.setWindowTitle('Scenario Image Automation')
        # Make window smaller and more compact
        min_width = 640
        min_height = 640
        self.setMinimumSize(min_width, min_height)
        self.setGeometry(100, 100, min_width, min_height)
        self.running = False
        self.hotkey = '<f9>'
        self.listener = None
        self.worker = None
        self.current_scenario = None
        self.selected_step_idx = None
        self.last_toggle_time = 0  # For debounce of start/stop
        
        # Memory optimization attributes
        self._last_screenshot = None
        self._last_screenshot_time = 0
        self._screenshot_cache_duration = 0.05  # Cache screenshot for 50ms
        self._processing_step = False
        self._step_cooldown = {}  # Cooldown tracking for steps
        
        # Performance monitoring
        self._loop_times = []
        self._max_loop_time_samples = 10
        self._last_warning_messages = {}  # Track last warning messages to prevent duplicates
        self._warning_cooldown = 30  # Seconds between duplicate warnings
        
        self.init_ui()
        self.load_scenarios()
        
        # Restore window geometry (size and position)
        self._restore_window_geometry()
        
        # Setup cleanup on close
        self.setAttribute(QtCore.Qt.WidgetAttribute.WA_DeleteOnClose)
        
        # Setup periodic monitoring timer
        self.monitor_timer = QtCore.QTimer()
        self.monitor_timer.timeout.connect(self._monitor_performance)
        
        # Get initial interval from dropdown (default is 2 seconds)
        initial_interval = self.update_interval_combo.currentData() or 2000
        self.monitor_timer.start(initial_interval)
        
        # Initial resource display update
        QtCore.QTimer.singleShot(100, self._monitor_performance)
    
    def _monitor_performance(self):
        """
        Monitor application performance and memory usage.
        """
        try:
            memory_info = self.get_memory_usage()
            
            # Update resource display
            self._update_resource_display(memory_info)
            
            # Log performance stats periodically (only if running)
            if self.running and hasattr(self, '_loop_times') and self._loop_times:
                avg_loop_time = sum(self._loop_times) / len(self._loop_times)
                
                # Warning thresholds with duplicate prevention
                if avg_loop_time > 1.0:
                    warning_msg = f"Performance warning: Average loop time {avg_loop_time:.3f}s"
                    self._log_warning_once("performance_slow", warning_msg)
                
                # Memory warning for cache-based monitoring
                if memory_info.get('template_cache_size', 0) > 20:
                    cache_msg = f"Template cache has {memory_info['template_cache_size']} entries"
                    self._log_warning_once("cache_large", cache_msg)
                
                # High memory usage warning
                if memory_info.get('process_memory_percent', 0) > 15:
                    memory_msg = f"High memory usage: {memory_info['process_memory_percent']:.1f}% of system memory"
                    self._log_warning_once("memory_high", memory_msg)
                
                # High CPU usage warning
                if memory_info.get('cpu_percent', 0) > 70:
                    cpu_msg = f"High CPU usage: {memory_info['cpu_percent']:.1f}%"
                    self._log_warning_once("cpu_high", cpu_msg)
                
                # GPU memory usage warning
                if memory_info.get('gpu_has_gpu', False) and memory_info.get('gpu_utilization_percent', 0) > 90:
                    gpu_msg = f"High GPU memory usage: {memory_info['gpu_utilization_percent']:.1f}%"
                    self._log_warning_once("gpu_high", gpu_msg)
                
        except Exception as e:
            logger.debug(f"Performance monitoring error: {e}")
    
    def _log_warning_once(self, warning_type, message):
        """
        Log a warning message only once per cooldown period to prevent spam.
        """
        current_time = time.time()
        
        # Check if we've already logged this type of warning recently
        if warning_type in self._last_warning_messages:
            last_time, last_message = self._last_warning_messages[warning_type]
            
            # If the message is the same and within cooldown period, skip
            if (current_time - last_time < self._warning_cooldown and 
                message == last_message):
                return
        
        # Log the warning and update tracking
        logger.warning(message)
        self._last_warning_messages[warning_type] = (current_time, message)
    
    def _update_resource_display(self, memory_info):
        """
        Update the resource usage display in the UI.
        """
        try:
            # Memory usage
            if memory_info.get('has_psutil', False):
                memory_text = f"Memory: {memory_info['process_memory_mb']:.1f}MB ({memory_info['process_memory_percent']:.1f}%)"
                memory_color = '#d9534f' if memory_info['process_memory_percent'] > 10 else '#5cb85c'
            else:
                memory_text = "Memory: N/A (psutil needed)"
                memory_color = '#f0ad4e'
            
            self.memory_label.setText(memory_text)
            self.memory_label.setStyleSheet(f'font-size: 9pt; color: {memory_color};')
            
            # Make memory label clickable to show psutil installation info
            if not memory_info.get('has_psutil', False):
                self.memory_label.mousePressEvent = lambda event: self._show_psutil_info()
                self.memory_label.setCursor(QtGui.QCursor(QtCore.Qt.CursorShape.PointingHandCursor))
            
            # CPU usage
            if memory_info.get('has_psutil', False):
                cpu_text = f"CPU: {memory_info['cpu_percent']:.1f}%"
                cpu_color = '#d9534f' if memory_info['cpu_percent'] > 50 else '#5cb85c'
            else:
                cpu_text = "CPU: N/A"
                cpu_color = '#777'
            
            self.cpu_label.setText(cpu_text)
            self.cpu_label.setStyleSheet(f'font-size: 9pt; color: {cpu_color};')
            
            # System memory
            if memory_info.get('has_psutil', False):
                system_text = f"System: {memory_info['system_memory_percent']:.1f}% ({memory_info['system_memory_available_gb']:.1f}GB free)"
                system_color = '#d9534f' if memory_info['system_memory_percent'] > 85 else '#5cb85c'
            else:
                system_text = "System: N/A"
                system_color = '#777'
            
            self.system_memory_label.setText(system_text)
            self.system_memory_label.setStyleSheet(f'font-size: 9pt; color: {system_color};')
            
            # Cache info
            cache_text = f"Cache: {memory_info['template_cache_size']} templates, {memory_info['cooldown_entries']} cooldowns"
            cache_color = '#f0ad4e' if memory_info['template_cache_size'] > 20 else '#5bc0de'
            
            self.cache_label.setText(cache_text)
            self.cache_label.setStyleSheet(f'font-size: 9pt; color: {cache_color};')
            
            # GPU memory info
            if memory_info.get('gpu_has_gpu', False):
                gpu_used_mb = memory_info.get('gpu_used_mb', 0)
                gpu_total_mb = memory_info.get('gpu_total_mb', 0)
                gpu_utilization = memory_info.get('gpu_utilization_percent', 0)
                gpu_method = memory_info.get('gpu_method', 'unknown')
                
                if gpu_total_mb > 0:
                    gpu_text = f"GPU: {gpu_used_mb:.0f}/{gpu_total_mb:.0f}MB ({gpu_utilization:.1f}%)"
                    gpu_color = '#d9534f' if gpu_utilization > 80 else '#f0ad4e' if gpu_utilization > 60 else '#5cb85c'
                    
                    # Add GPU name as tooltip
                    gpu_name = memory_info.get('gpu_name', 'Unknown GPU')
                    self.gpu_label.setToolTip(f"GPU: {gpu_name} (detected via {gpu_method})")
                else:
                    gpu_text = f"GPU: Detected ({gpu_method})"
                    gpu_color = '#5bc0de'
                    gpu_name = memory_info.get('gpu_name', 'Unknown GPU')
                    self.gpu_label.setToolTip(f"GPU: {gpu_name} (limited info via {gpu_method})")
            else:
                gpu_text = "GPU: Not detected"
                gpu_color = '#777'
                self.gpu_label.setToolTip(
                    "No GPU detected or GPU monitoring libraries not available.\n\n"
                    "For enhanced GPU monitoring:\n"
                    "• NVIDIA GPUs: pip install pynvml or GPUtil\n"
                    "• Integrated GPUs: Built-in Windows WMI support\n"
                    "• Linux: lspci and DRM detection\n"
                    "• macOS: system_profiler integration\n\n"
                    "Click for more information about GPU monitoring."
                )
            
            self.gpu_label.setText(gpu_text)
            self.gpu_label.setStyleSheet(f'font-size: 9pt; color: {gpu_color};')
            
            # Make GPU label clickable to show GPU info when no GPU detected
            if not memory_info.get('gpu_has_gpu', False):
                self.gpu_label.mousePressEvent = lambda event: self._show_gpu_info()
                self.gpu_label.setCursor(QtGui.QCursor(QtCore.Qt.CursorShape.PointingHandCursor))
            else:
                # Remove click handler if GPU is detected
                self.gpu_label.mousePressEvent = None
                self.gpu_label.setCursor(QtGui.QCursor(QtCore.Qt.CursorShape.ArrowCursor))
            
            # Performance info
            if hasattr(self, '_loop_times') and self._loop_times:
                avg_time = sum(self._loop_times) / len(self._loop_times)
                perf_text = f"Performance: {avg_time*1000:.0f}ms avg loop time"
                perf_color = '#d9534f' if avg_time > 0.5 else '#5cb85c'
            else:
                perf_text = "Performance: Not running"
                perf_color = '#777'
            
            self.performance_label.setText(perf_text)
            self.performance_label.setStyleSheet(f'font-size: 9pt; color: {perf_color};')
            
            # Show error if any
            if 'error' in memory_info:
                self.performance_label.setText(f"Error: {memory_info['error'][:50]}...")
                self.performance_label.setStyleSheet('font-size: 9pt; color: #d9534f;')
                
        except Exception as e:
            logger.debug(f"Error updating resource display: {e}")
            # Show basic fallback info
            self.memory_label.setText("Memory: Error")
            self.cpu_label.setText("CPU: Error")
            self.system_memory_label.setText("System: Error")
            self.cache_label.setText("Cache: Error")
            self.gpu_label.setText("GPU: Error")
            self.performance_label.setText(f"Display Error: {str(e)[:30]}...")
    
    def stop_monitoring(self):
        """Stop performance monitoring timer."""
        if hasattr(self, 'monitor_timer') and self.monitor_timer:
            self.monitor_timer.stop()
    
    def _toggle_resource_display(self):
        """Toggle the visibility of resource usage widgets."""
        if self.resource_widgets_visible:
            # Hide resource widgets
            self.memory_label.hide()
            self.cpu_label.hide()
            self.system_memory_label.hide()
            self.cache_label.hide()
            self.gpu_label.hide()
            self.performance_label.hide()
            self.interval_label.hide()
            self.update_interval_combo.hide()
            self.toggle_resources_btn.setText('Show Resources')
            self.resource_widgets_visible = False
        else:
            # Show resource widgets
            self.memory_label.show()
            self.cpu_label.show()
            self.system_memory_label.show()
            self.cache_label.show()
            self.gpu_label.show()
            self.performance_label.show()
            self.interval_label.show()
            self.update_interval_combo.show()
            self.toggle_resources_btn.setText('Hide Resources')
            self.resource_widgets_visible = True
    
    def _show_psutil_info(self):
        """Show information about installing psutil for enhanced monitoring."""
        msg = QtWidgets.QMessageBox(self)
        msg.setWindowTitle("Enhanced Resource Monitoring")
        msg.setIcon(QtWidgets.QMessageBox.Icon.Information)
        msg.setText("Install psutil for detailed resource monitoring")
        msg.setInformativeText(
            "For detailed CPU, memory, and system resource monitoring, install the psutil package:\n\n"
            "pip install psutil\n\n"
            "This will enable:\n"
            "• Process memory usage in MB and percentage\n"
            "• CPU usage percentage\n"
            "• System memory statistics\n"
            "• Available system memory\n\n"
            "Without psutil, only basic cache information is shown."
        )
        msg.setStandardButtons(QtWidgets.QMessageBox.StandardButton.Ok)
        msg.exec()
    
    def _show_gpu_info(self):
        """Show information about installing GPU monitoring libraries."""
        msg = QtWidgets.QMessageBox(self)
        msg.setWindowTitle("GPU Memory Monitoring")
        msg.setIcon(QtWidgets.QMessageBox.Icon.Information)
        msg.setText("GPU monitoring supports both discrete and integrated GPUs")
        msg.setInformativeText(
            "GPU Memory Monitoring Support:\n\n"
            "NVIDIA GPUs (Discrete):\n"
            "• pip install pynvml (recommended)\n"
            "• pip install GPUtil (alternative)\n"
            "• nvidia-smi command line tool\n\n"
            "Integrated GPUs (Intel/AMD):\n"
            "• Windows: WMI queries (built-in)\n"
            "• Linux: lspci and /sys/class/drm\n"
            "• macOS: system_profiler\n\n"
            "Features enabled:\n"
            "• GPU detection and identification\n"
            "• VRAM/memory size (where available)\n"
            "• Real-time usage (NVIDIA with libraries)\n"
            "• Cross-platform support\n\n"
            "Note: Integrated GPUs may show limited information\n"
            "compared to discrete GPUs due to system limitations.\n"
            "NVIDIA GPUs with proper libraries provide the most\n"
            "detailed monitoring including real-time usage."
        )
        msg.setStandardButtons(QtWidgets.QMessageBox.StandardButton.Ok)
        msg.exec()
    
    def _on_update_interval_changed(self):
        """Handle resource monitoring update interval change."""
        interval = self.update_interval_combo.currentData()
        if interval and hasattr(self, 'monitor_timer'):
            # Stop current timer
            self.monitor_timer.stop()
            
            # Start timer with new interval
            self.monitor_timer.start(interval)
            
            logger.debug(f"Resource monitoring interval changed to {interval}ms")
            
            # Immediately update display to show the change is active
            QtCore.QTimer.singleShot(50, self._monitor_performance)
    
    def _save_window_geometry(self):
        """Save the current window size and position to a config file."""
        try:
            config_path = os.path.join(CONFIG_DIR, 'window_config.json')
            geometry_data = {
                'x': self.x(),
                'y': self.y(),
                'width': self.width(),
                'height': self.height(),
                'maximized': self.isMaximized()
            }
            
            # Load existing config if it exists
            existing_config = {}
            if os.path.exists(config_path):
                try:
                    with open(config_path, 'r') as f:
                        existing_config = json.load(f)
                except (json.JSONDecodeError, IOError):
                    existing_config = {}
            
            # Update with window geometry and settings
            existing_config['main_window'] = geometry_data
            existing_config['update_interval'] = self.update_interval_combo.currentData()
            
            with open(config_path, 'w') as f:
                json.dump(existing_config, f, indent=2)
            
            logger.debug(f"Saved window geometry and settings: {geometry_data}")
            
        except Exception as e:
            logger.debug(f"Error saving window geometry: {e}")
    
    def _restore_window_geometry(self):
        """Restore the window size and position from config file."""
        try:
            config_path = os.path.join(CONFIG_DIR, 'window_config.json')
            
            if not os.path.exists(config_path):
                logger.debug("No window config file found, using default geometry")
                return
            
            with open(config_path, 'r') as f:
                config = json.load(f)
            
            geometry_data = config.get('main_window')
            if not geometry_data:
                logger.debug("No main window geometry data found")
                return
            
            # Validate geometry data
            required_keys = ['x', 'y', 'width', 'height']
            if not all(key in geometry_data for key in required_keys):
                logger.debug("Incomplete geometry data, using defaults")
                return
            
            # Get screen dimensions to validate position
            screen = QtWidgets.QApplication.primaryScreen().geometry()
            screen_width = screen.width()
            screen_height = screen.height()
            
            # Validate and adjust position if necessary
            x = max(0, min(geometry_data['x'], screen_width - 100))  # Ensure at least 100px visible
            y = max(0, min(geometry_data['y'], screen_height - 100))
            width = max(640, min(geometry_data['width'], screen_width))  # Minimum 640px wide
            height = max(480, min(geometry_data['height'], screen_height))  # Minimum 480px tall
            
            # Apply geometry
            self.setGeometry(x, y, width, height)
            
            # Restore maximized state if applicable
            if geometry_data.get('maximized', False):
                self.showMaximized()
            
            # Restore update interval if saved
            saved_interval = config.get('update_interval')
            if saved_interval:
                # Find the index of the saved interval in the combo box
                for i in range(self.update_interval_combo.count()):
                    if self.update_interval_combo.itemData(i) == saved_interval:
                        self.update_interval_combo.setCurrentIndex(i)
                        break
                logger.debug(f"Restored update interval: {saved_interval}ms")
            
            logger.debug(f"Restored window geometry: x={x}, y={y}, w={width}, h={height}")
            
        except (json.JSONDecodeError, IOError, KeyError) as e:
            logger.debug(f"Error loading window geometry: {e}")
        except Exception as e:
            logger.warning(f"Unexpected error restoring window geometry: {e}")
    
    def moveEvent(self, event):
        """Handle window move events to save position."""
        super().moveEvent(event)
        # Save geometry after a short delay to avoid excessive writes during dragging
        if hasattr(self, '_geometry_save_timer'):
            self._geometry_save_timer.stop()
        else:
            self._geometry_save_timer = QtCore.QTimer()
            self._geometry_save_timer.setSingleShot(True)
            self._geometry_save_timer.timeout.connect(self._save_window_geometry)
        
        self._geometry_save_timer.start(500)  # Save after 500ms of no movement
    
    def resizeEvent(self, event):
        """Handle window resize events to save size."""
        super().resizeEvent(event)
        # Save geometry after a short delay to avoid excessive writes during resizing
        if hasattr(self, '_geometry_save_timer'):
            self._geometry_save_timer.stop()
        else:
            self._geometry_save_timer = QtCore.QTimer()
            self._geometry_save_timer.setSingleShot(True)
            self._geometry_save_timer.timeout.connect(self._save_window_geometry)
        
        self._geometry_save_timer.start(500)  # Save after 500ms of no resizing
    
    def changeEvent(self, event):
        """Handle window state changes (maximize/minimize)."""
        super().changeEvent(event)
        if event.type() == QtCore.QEvent.Type.WindowStateChange:
            # Save geometry when window state changes
            QtCore.QTimer.singleShot(100, self._save_window_geometry)
    
    def closeEvent(self, event):
        """Handle application close event with proper cleanup."""
        # Save window geometry before closing
        self._save_window_geometry()
        
        self.cleanup_resources()
        event.accept()
    
    def cleanup_resources(self):
        """Clean up resources to prevent memory leaks."""
        if self.running:
            self.stop_automation()
        
        # Stop monitoring timer
        self.stop_monitoring()
        
        # Clear template cache
        template_cache.clear()
        
        # Clear screenshot cache
        self._last_screenshot = None
        
        # Clear performance tracking
        if hasattr(self, '_loop_times'):
            self._loop_times.clear()
        if hasattr(self, '_step_cooldown'):
            self._step_cooldown.clear()
        
        # Force garbage collection
        gc.collect()
        
        logger.info("Resources cleaned up successfully")

    def init_ui(self):
        """
        Initializes the main UI components and layouts.
        Sets up scenario and window dropdowns, step list, and action buttons.
        """
        # Set compact font and style
        font = QtGui.QFont()
        font.setPointSize(9)
        self.setFont(font)
        style = """
            QPushButton, QComboBox, QListWidget, QLabel {
                font-size: 9pt;
                min-height: 20px;
                padding: 2px 6px;
            }
            QComboBox { min-width: 90px; }
            QPushButton { min-width: 70px; }
            QListWidget { min-width: 200px; min-height: 120px; }
        """
        self.setStyleSheet(style)

        # Scenario group (QGroupBox with title 'Scenario')
        scenario_group_box = QtWidgets.QGroupBox('Scenario')
        scenario_group_box.setSizePolicy(QtWidgets.QSizePolicy.Policy.Preferred, QtWidgets.QSizePolicy.Policy.Fixed)
        scenario_group_layout = QtWidgets.QVBoxLayout()
        self.combo = QtWidgets.QComboBox()
        self.combo.setMinimumWidth(90)
        self.combo.currentIndexChanged.connect(self._log_combo_scenario)
        scenario_group_layout.addWidget(self.combo)

        # Window selection dropdown
        self.window_combo = QtWidgets.QComboBox()
        self.window_combo.setMinimumWidth(90)
        # --- Window selection group ---
        window_group_box = QtWidgets.QGroupBox('Target Window')
        window_group_box.setSizePolicy(QtWidgets.QSizePolicy.Policy.Preferred, QtWidgets.QSizePolicy.Policy.Fixed)
        window_group_layout = QtWidgets.QHBoxLayout()
        self.window_combo.setMinimumWidth(90)
        self.window_combo.currentIndexChanged.connect(self._log_combo_window)
        self.btn_refresh_windows = QtWidgets.QPushButton('Refresh')
        self.btn_refresh_windows.setToolTip('Refresh open windows list')
        self.btn_refresh_windows.clicked.connect(self._log_btn_refresh_windows)
        window_group_layout.addWidget(self.window_combo)
        window_group_layout.addWidget(self.btn_refresh_windows)
        window_group_box.setLayout(window_group_layout)
        self.refresh_window_list()
        # Scenario actions
        self.btn_new = QtWidgets.QPushButton('New')
        self.btn_new.setToolTip('Create new scenario')
        self.btn_new.clicked.connect(self._log_btn_new)
        self.btn_import = QtWidgets.QPushButton('Import')
        self.btn_import.setToolTip('Import scenario')
        self.btn_import.clicked.connect(self._log_btn_import)
        self.btn_export = QtWidgets.QPushButton('Export')
        self.btn_export.setToolTip('Export scenario')
        self.btn_export.clicked.connect(self._log_btn_export)
        self.btn_rename_scenario = QtWidgets.QPushButton("Rename")
        self.btn_rename_scenario.setToolTip('Rename scenario')
        self.btn_rename_scenario.clicked.connect(self._log_btn_rename_scenario)
        self.btn_delete_scenario = QtWidgets.QPushButton("Delete")
        self.btn_delete_scenario.setToolTip('Delete current scenario')
        self.btn_delete_scenario.clicked.connect(self._log_btn_delete_scenario)
        scenario_btn_group_layout = QtWidgets.QHBoxLayout()
        scenario_btn_group_layout.setSpacing(4)
        scenario_btn_group_layout.setContentsMargins(4, 4, 4, 4)
        scenario_btn_group_layout.addWidget(self.btn_new)
        scenario_btn_group_layout.addWidget(self.btn_import)
        scenario_btn_group_layout.addWidget(self.btn_export)
        scenario_btn_group_layout.addWidget(self.btn_rename_scenario)
        scenario_btn_group_layout.addWidget(self.btn_delete_scenario)
        scenario_group_layout.addLayout(scenario_btn_group_layout)
        scenario_group_box.setLayout(scenario_group_layout)

        # Steps group (QGroupBox with title 'Steps')
        steps_group_box = QtWidgets.QGroupBox('Steps (Up = Higher Priority, Down = Lower Priority)')
        steps_group_layout = QtWidgets.QVBoxLayout()
        self.steps_list = QtWidgets.QListWidget()
        self.steps_list.setMinimumHeight(120)
        self.steps_list.currentRowChanged.connect(self._log_steps_list)
        steps_group_layout.addWidget(self.steps_list)
        # Step actions
        self.btn_add_step = QtWidgets.QPushButton('Add')
        self.btn_add_step.setToolTip('Add step')
        self.btn_add_step.clicked.connect(self._log_btn_add_step)
        self.btn_edit_step = QtWidgets.QPushButton('Edit')
        self.btn_edit_step.setToolTip('Edit step')
        self.btn_edit_step.clicked.connect(self._log_btn_edit_step)
        self.btn_del_step = QtWidgets.QPushButton('Delete')
        self.btn_del_step.setToolTip('Delete step')
        self.btn_del_step.clicked.connect(self._log_btn_del_step)
        self.btn_rename_step = QtWidgets.QPushButton("Rename")
        self.btn_rename_step.setToolTip('Rename step')
        self.btn_rename_step.clicked.connect(self._log_btn_rename_step)
        self.btn_move_up_step = QtWidgets.QPushButton('Move Up')
        self.btn_move_up_step.setToolTip('Move step up (higher priority)')
        self.btn_move_up_step.clicked.connect(self._log_btn_move_up_step)
        self.btn_move_down_step = QtWidgets.QPushButton('Move Down')
        self.btn_move_down_step.setToolTip('Move step down (lower priority)')
        self.btn_move_down_step.clicked.connect(self._log_btn_move_down_step)
        step_btn_group_layout = QtWidgets.QHBoxLayout()
        step_btn_group_layout.setSpacing(4)
        step_btn_group_layout.setContentsMargins(4, 4, 4, 4)
        step_btn_group_layout.addWidget(self.btn_add_step)
        step_btn_group_layout.addWidget(self.btn_edit_step)
        step_btn_group_layout.addWidget(self.btn_del_step)
        step_btn_group_layout.addWidget(self.btn_rename_step)
        step_btn_group_layout.addWidget(self.btn_move_up_step)
        step_btn_group_layout.addWidget(self.btn_move_down_step)
        steps_group_layout.addLayout(step_btn_group_layout)
        steps_group_box.setLayout(steps_group_layout)

        # Start/Stop Combined
        self.btn_start_stop = QtWidgets.QPushButton(f'Start ({self.hotkey.upper()})')
        self.btn_start_stop.setMinimumWidth(80)
        self.btn_start_stop.clicked.connect(self._log_btn_start_stop)

        # State label
        self.state_label = QtWidgets.QLabel('State: Paused')
        self.state_label.setStyleSheet('font-weight: bold; font-size: 11pt; color: #0055aa;')
        self.state_label.setAlignment(QtCore.Qt.AlignmentFlag.AlignLeft | QtCore.Qt.AlignmentFlag.AlignVCenter)

        # Resource usage display
        self.resource_group = QtWidgets.QGroupBox('Resource Usage')
        self.resource_group.setSizePolicy(QtWidgets.QSizePolicy.Policy.Preferred, QtWidgets.QSizePolicy.Policy.Fixed)
        resource_layout = QtWidgets.QGridLayout()
        resource_layout.setSpacing(4)
        resource_layout.setContentsMargins(6, 6, 6, 6)

        # Toggle button for resource display
        self.toggle_resources_btn = QtWidgets.QPushButton('Hide Resources')
        self.toggle_resources_btn.setMaximumWidth(100)
        self.toggle_resources_btn.setToolTip('Toggle resource usage display. Install psutil for detailed system metrics.')
        self.toggle_resources_btn.clicked.connect(self._toggle_resource_display)
        resource_layout.addWidget(self.toggle_resources_btn, 0, 2, 1, 1)

        # Update interval dropdown
        self.interval_label = QtWidgets.QLabel('Update:')
        self.interval_label.setStyleSheet('font-size: 9pt; color: #333;')
        self.update_interval_combo = QtWidgets.QComboBox()
        self.update_interval_combo.setMaximumWidth(80)
        self.update_interval_combo.setToolTip('Set resource monitoring update interval')
        
        # Add interval options (in milliseconds)
        intervals = [
            ('0.5s', 500),
            ('1s', 1000),
            ('2s', 2000),
            ('3s', 3000),
            ('5s', 5000),
            ('10s', 10000)
        ]
        
        for text, value in intervals:
            self.update_interval_combo.addItem(text, value)
        
        # Set default to 2 seconds (index 2)
        self.update_interval_combo.setCurrentIndex(2)
        self.update_interval_combo.currentIndexChanged.connect(self._on_update_interval_changed)
        
        resource_layout.addWidget(self.interval_label, 0, 3, 1, 1)
        resource_layout.addWidget(self.update_interval_combo, 0, 4, 1, 1)

        # Memory usage
        self.memory_label = QtWidgets.QLabel('Memory: --')
        self.memory_label.setStyleSheet('font-size: 9pt; color: #333;')
        resource_layout.addWidget(self.memory_label, 0, 0)

        # CPU usage
        self.cpu_label = QtWidgets.QLabel('CPU: --')
        self.cpu_label.setStyleSheet('font-size: 9pt; color: #333;')
        resource_layout.addWidget(self.cpu_label, 0, 1)

        # System memory
        self.system_memory_label = QtWidgets.QLabel('System: --')
        self.system_memory_label.setStyleSheet('font-size: 9pt; color: #333;')
        resource_layout.addWidget(self.system_memory_label, 1, 0)

        # Cache info
        self.cache_label = QtWidgets.QLabel('Cache: --')
        self.cache_label.setStyleSheet('font-size: 9pt; color: #333;')
        resource_layout.addWidget(self.cache_label, 1, 1)

        # GPU memory info
        self.gpu_label = QtWidgets.QLabel('GPU: --')
        self.gpu_label.setStyleSheet('font-size: 9pt; color: #333;')
        resource_layout.addWidget(self.gpu_label, 1, 2)

        # Performance info
        self.performance_label = QtWidgets.QLabel('Performance: --')
        self.performance_label.setStyleSheet('font-size: 9pt; color: #333;')
        resource_layout.addWidget(self.performance_label, 2, 0, 1, 3)

        self.resource_group.setLayout(resource_layout)
        self.resource_widgets_visible = True

        # Main layout
        layout = QtWidgets.QGridLayout()
        layout.setHorizontalSpacing(6)
        layout.setVerticalSpacing(4)
        layout.setContentsMargins(8, 8, 8, 8)

        # Scenario, Window, and Steps groups in layout
        layout.addWidget(scenario_group_box, 0, 0, 1, 5)
        layout.setRowStretch(0, 0)  # Scenario group should not stretch vertically
        layout.addWidget(window_group_box, 1, 0, 1, 5)
        layout.addWidget(steps_group_box, 2, 0, 3, 5)

        # Resource usage group
        layout.addWidget(self.resource_group, 5, 0, 1, 5)
        layout.setRowStretch(5, 0)  # Resource group should not stretch vertically

        # Start/State row in a group
        start_state_group = QtWidgets.QGroupBox()
        start_state_group.setTitle("")
        start_state_group.setSizePolicy(QtWidgets.QSizePolicy.Policy.Preferred, QtWidgets.QSizePolicy.Policy.Fixed)
        start_state_layout = QtWidgets.QHBoxLayout()
        start_state_layout.setSpacing(6)  # Fixed gap between button and label
        start_state_layout.addWidget(self.btn_start_stop)
        start_state_layout.addWidget(self.state_label)
        start_state_layout.addStretch(1)
        start_state_group.setLayout(start_state_layout)
        layout.addWidget(start_state_group, 6, 0, 1, 5)
        layout.setRowStretch(6, 0)  # Prevent vertical stretch

        # Set central widget (must be at the end of init_ui)
        central = QtWidgets.QWidget()
        central.setLayout(layout)
        self.setCentralWidget(central)

    def _log_btn_refresh_windows(self):
        logger.info("UI: Refresh Windows button pressed")
        self.refresh_window_list()

    def _log_combo_scenario(self, idx):
        """
        Log scenario combo box change and call select_scenario.
        """
        logger.info(f"UI: Scenario combo changed to index {idx} ({self.combo.currentText()})")
        self.select_scenario()

    def _log_combo_window(self, idx):
        """
        Log window combo box change and call save_selected_window.
        """
        logger.info(f"UI: Window combo changed to index {idx} ({self.window_combo.currentText()})")
        self.save_selected_window()

    def _log_btn_new(self):
        """
        Log New Scenario button press and call create_scenario.
        """
        logger.info("UI: New Scenario button pressed")
        self.create_scenario()

    def _log_btn_import(self):
        """
        Log Import Scenario button press and call import_scenario.
        """
        logger.info("UI: Import Scenario button pressed")
        self.import_scenario()

    def _log_btn_export(self):
        """
        Log Export Scenario button press and call export_scenario.
        """
        logger.info("UI: Export Scenario button pressed")
        self.export_scenario()

    def _log_btn_rename_scenario(self):
        """
        Log Rename Scenario button press and call rename_scenario.
        """
        logger.info("UI: Rename Scenario button pressed")
        self.rename_scenario()

    def _log_btn_delete_scenario(self):
        """
        Log Delete Scenario button press and call delete_scenario.
        """
        logger.info("UI: Delete Scenario button pressed")
        self.delete_scenario()

    def delete_scenario(self):
        """
        Delete the current scenario after confirmation.
        """
        if not self.current_scenario:
            return
        name = self.current_scenario.name
        reply = QtWidgets.QMessageBox.question(
            self,
            "Delete Scenario",
            f"Are you sure you want to delete the scenario '{name}'? This cannot be undone.",
            QtWidgets.QMessageBox.StandardButton.Yes | QtWidgets.QMessageBox.StandardButton.No
        )
        if reply == QtWidgets.QMessageBox.StandardButton.Yes:
            try:
                scenario_dir = self.current_scenario.get_scenario_dir()
                if os.path.exists(scenario_dir):
                    shutil.rmtree(scenario_dir)
                logger.info(f"Deleted scenario: {name}")
                self.current_scenario = None
                self.load_scenarios()
            except Exception as e:
                logger.error(f"Failed to delete scenario '{name}': {e}")
                QtWidgets.QMessageBox.critical(self, "Delete Error", f"Failed to delete scenario.\n{e}")

    def _log_steps_list(self, idx):
        """
        Log steps list row change and call select_step.
        """
        logger.info(f"UI: Steps list row changed to {idx}")
        self.select_step(idx)

    def _log_btn_add_step(self):
        """
        Log Add Step button press and call add_step.
        """
        logger.info("UI: Add Step button pressed")
        self.add_step()

    def _log_btn_edit_step(self):
        """
        Log Edit Step button press and call edit_step.
        """
        logger.info("UI: Edit Step button pressed")
        self.edit_step()

    def _log_btn_del_step(self):
        """
        Log Delete Step button press and call delete_step.
        """
        logger.info("UI: Delete Step button pressed")
        self.delete_step()

    def _log_btn_rename_step(self):
        """
        Log Rename Step button press and call rename_step.
        """
        logger.info("UI: Rename Step button pressed")
        self.rename_step()

    def _log_btn_start_stop(self):
        """
        Log Start/Stop button press and call toggle_automation.
        """
        logger.info("UI: Start/Stop button pressed")
        self.toggle_automation()

    def _log_btn_move_up_step(self):
        """
        Log Move Up Step button press and call move_up_step.
        """
        logger.info("UI: Move Up Step button pressed")
        self.move_up_step()

    def _log_btn_move_down_step(self):
        """
        Log Move Down Step button press and call move_down_step.
        """
        logger.info("UI: Move Down Step button pressed")
        self.move_down_step()

    def move_up_step(self):
        """
        Move the selected step up in the steps list (higher priority).
        """
        idx = self.selected_step_idx
        if (
            self.current_scenario
            and idx is not None
            and idx > 0
            and idx < len(self.current_scenario.steps)
        ):
            self.current_scenario.steps[idx - 1], self.current_scenario.steps[idx] = (
                self.current_scenario.steps[idx],
                self.current_scenario.steps[idx - 1],
            )
            self.current_scenario.save()
            self.refresh_lists()
            self.steps_list.setCurrentRow(idx - 1)
            self.selected_step_idx = idx - 1

    def move_down_step(self):
        """
        Move the selected step down in the steps list (lower priority).
        """
        idx = self.selected_step_idx
        if (
            self.current_scenario
            and idx is not None
            and idx >= 0
            and idx < len(self.current_scenario.steps) - 1
        ):
            self.current_scenario.steps[idx + 1], self.current_scenario.steps[idx] = (
                self.current_scenario.steps[idx],
                self.current_scenario.steps[idx + 1],
            )
            self.current_scenario.save()
            self.refresh_lists()
            self.steps_list.setCurrentRow(idx + 1)
            self.selected_step_idx = idx + 1

    def select_step(self, idx):
        """
        Set the selected step index in the UI.
        """
        self.selected_step_idx = idx

    def add_step(self):
        """
        Open the dialog to add a new step to the current scenario.
        """
        self._step_dialog = StepDialog(self)
        self._step_dialog.accepted.connect(self._on_add_step_accepted)
        self._step_dialog.show()

    def _on_add_step_accepted(self):
        step = self._step_dialog.get_step()
        self.current_scenario.steps.append(step)
        self.current_scenario.save()
        self.refresh_lists()
        logger.info(f'Added step: {step}')
        self._step_dialog.deleteLater()
        self._step_dialog = None

    def edit_step(self):
        """
        Open the dialog to edit the selected step in the current scenario.
        """
        idx = self.selected_step_idx
        if idx is None or idx < 0 or idx >= len(self.current_scenario.steps):
            return
        step = self.current_scenario.steps[idx]
        self._step_dialog = StepDialog(self, step)
        self._step_dialog.accepted.connect(lambda: self._on_edit_step_accepted(idx))
        self._step_dialog.show()

    def _on_edit_step_accepted(self, idx):
        self.current_scenario.steps[idx] = self._step_dialog.get_step()
        self.current_scenario.save()
        self.refresh_lists()
        logger.info(f'Edited step at idx {idx}')
        self._step_dialog.deleteLater()
        self._step_dialog = None

    def delete_step(self):
        """
        Delete the selected step from the current scenario.
        """
        idx = self.selected_step_idx
        if idx is None or idx < 0 or idx >= len(self.current_scenario.steps):
            return
        del self.current_scenario.steps[idx]
        self.current_scenario.save()
        self.refresh_lists()
        logger.info(f'Deleted step at idx {idx}')

    def save_last_scenario(self, name):
        """
        Save the name of the last selected scenario to disk.
        """
        try:
            with open(os.path.join(CONFIG_DIR, 'last_scenario.txt'), 'w') as f:
                f.write(name)
        except Exception as e:
            logger.error(f"Could not save last scenario: {e}")

    def read_last_scenario(self):
        """
        Read the name of the last selected scenario from disk.
        """
        try:
            path = os.path.join(CONFIG_DIR, 'last_scenario.txt')
            if not os.path.exists(path):
                return None
            with open(path, 'r') as f:
                return f.read().strip()
        except Exception as e:
            logger.error(f"Could not read last scenario: {e}")
            return None

    def load_scenarios(self):
        """
        Load all scenarios and populate the scenario combo box.
        """
        logger.debug('Loading scenarios...')
        self.combo.clear()
        scenarios = Scenario.list_all()
        for name in scenarios:
            self.combo.addItem(name)

        self.refresh_window_list()

        if not scenarios:
            choice_dialog = QtWidgets.QMessageBox(self)
            choice_dialog.setWindowTitle('No Scenarios Found')
            choice_dialog.setText('No scenarios found. Would you like to create a new scenario or import one?')
            create_btn = choice_dialog.addButton('Create New', QtWidgets.QMessageBox.ButtonRole.AcceptRole)
            import_btn = choice_dialog.addButton('Import', QtWidgets.QMessageBox.ButtonRole.ActionRole)
            choice_dialog.setDefaultButton(create_btn)
            choice_dialog.exec()
            clicked = choice_dialog.clickedButton()
            if clicked == create_btn:
                name, ok = QtWidgets.QInputDialog.getText(self, 'New Scenario', 'Enter a name for your first scenario:')
                if ok and name:
                    logger.info(f'Creating new scenario: {name}')
                    s = Scenario(name)
                    s.save()
                    self.combo.addItem(name)
                    self.combo.setCurrentText(name)
                else:
                    self.refresh_lists()
            elif clicked == import_btn:
                self.import_scenario()
            else:
                self.refresh_lists()
        else:
            last_scenario_name = self.read_last_scenario()
            if last_scenario_name and last_scenario_name in scenarios:
                self.combo.setCurrentText(last_scenario_name)
            else:
                self.combo.setCurrentIndex(0)

    def select_scenario(self):
        """
        Load the selected scenario and update the UI.
        """
        name = self.combo.currentText()
        logger.info(f'Scenario selected: {name}')
        if name:
            self.current_scenario = Scenario.load(name)
            self.save_last_scenario(name)
            self.refresh_lists()
            self.load_selected_window_from_config()

    def create_scenario(self):
        """
        Open a dialog to create a new scenario.
        """
        name, ok = QtWidgets.QInputDialog.getText(self, 'New Scenario', 'Enter scenario name:')
        if ok and name:
            logger.info(f'Creating new scenario: {name}')
            s = Scenario(name)
            s.save()
            self.load_scenarios()
            self.combo.setCurrentText(name)

    def rename_scenario(self):
        """
        Open a dialog to rename the current scenario.
        """
        if not self.current_scenario:
            return
        
        old_name = self.current_scenario.name
        new_name, ok = QtWidgets.QInputDialog.getText(self, "Rename Scenario", "New name:", text=old_name)
        
        if ok and new_name and new_name != old_name:
            # Rename the scenario directory
            try:
                old_dir = os.path.join(CONFIG_DIR, old_name)
                new_dir = os.path.join(CONFIG_DIR, new_name)
                if not os.path.exists(old_dir):
                    raise FileNotFoundError(f"Scenario directory '{old_dir}' does not exist.")
                if os.path.exists(new_dir):
                    raise FileExistsError(f"A scenario named '{new_name}' already exists.")
                os.rename(old_dir, new_dir)
            except OSError as e:
                logger.error(f"Error renaming scenario directory: {e}")
                QtWidgets.QMessageBox.critical(self, "Error", f"Could not rename scenario directory.\n{e}")
                return

            # Update the scenario object and save it
            self.current_scenario.name = new_name
            self.current_scenario.save()
            
            # Refresh the UI
            self.load_scenarios()
            self.combo.setCurrentText(new_name)

    def rename_step(self):
        """
        Open a dialog to rename the selected step.
        """
        if not self.current_scenario or self.selected_step_idx is None:
            return

        step = self.current_scenario.steps[self.selected_step_idx]
        old_name = step.get('name', 'Unnamed Step')
        new_name, ok = QtWidgets.QInputDialog.getText(self, "Rename Step", "New name:", text=old_name)

        if ok and new_name and new_name != old_name:
            step['name'] = new_name
            self.current_scenario.save()
            self.refresh_lists()

    def import_scenario(self):
        """
        Open a file dialog to import a scenario from a zip file.
        """
        path, _ = QtWidgets.QFileDialog.getOpenFileName(self, 'Import Scenario', '', 'ZIP Files (*.zip)')
        if path:
            try:
                with zipfile.ZipFile(path, 'r') as zip_ref:
                    # Extract to a temporary directory to inspect
                    temp_dir = os.path.join(CONFIG_DIR, "_temp_import")
                    if os.path.exists(temp_dir):
                        shutil.rmtree(temp_dir)
                    zip_ref.extractall(temp_dir)
                    
                    # Find the scenario name (the directory inside the temp folder)
                    extracted_dirs = [d for d in os.listdir(temp_dir) if os.path.isdir(os.path.join(temp_dir, d))]
                    if not extracted_dirs:
                        raise Exception("No scenario directory found in the zip file.")
                    scenario_name = extracted_dirs[0]
                    
                    # Move the extracted scenario to the main scenarios directory
                    src_path = os.path.join(temp_dir, scenario_name)
                    dest_path = os.path.join(CONFIG_DIR, scenario_name)
                    if os.path.exists(dest_path):
                        shutil.rmtree(dest_path)
                    shutil.move(src_path, dest_path)
                    
                    shutil.rmtree(temp_dir)

                logger.info(f'Imported scenario from {path}')
                self.load_scenarios()
                self.combo.setCurrentText(scenario_name)
            except Exception as e:
                logger.error(f'Failed to import scenario: {e}')
                QtWidgets.QMessageBox.critical(self, "Import Error", f"Failed to import scenario.\n{e}")

    def export_scenario(self):
        """
        Open a file dialog to export the current scenario to a zip file.
        """
        if not self.current_scenario:
            logger.warning('No scenario selected for export.')
            return

        scenario_dir = self.current_scenario.get_scenario_dir()
        path, _ = QtWidgets.QFileDialog.getSaveFileName(self, 'Export Scenario', f'{self.current_scenario.name}.zip', 'ZIP Files (*.zip)')

        if path:
            try:
                with zipfile.ZipFile(path, 'w', zipfile.ZIP_DEFLATED) as zipf:
                    for root, _, files in os.walk(scenario_dir):
                        for file in files:
                            file_path = os.path.join(root, file)
                            arcname = os.path.relpath(file_path, CONFIG_DIR)
                            zipf.write(file_path, arcname)
                logger.info(f'Exported scenario to {path}')
            except Exception as e:
                logger.error(f'Failed to export scenario: {e}')
                QtWidgets.QMessageBox.critical(self, "Export Error", f"Failed to export scenario.\n{e}")

    
    def add_action(self):
        """
        Open the dialog to add a new action to the step.
        """
        """
        Open the dialog to add an action to the current scenario.
        """
        logger.debug('Add Action: Opening action dialog.')
        dlg = ActionDialog(self, self.current_scenario.images)
        if dlg.exec():
            action = dlg.get_action()
            self.current_scenario.actions.append(action)
            self.current_scenario.save()
            self.refresh_lists()
            logger.info(f'Action added: {action}')

    def refresh_lists(self):
        """
        Refresh the step list widget with the current scenario's steps.
        """
        logger.debug('Refreshing steps list.')
        self.steps_list.clear()
        if not self.current_scenario:
            return
        for step in self.current_scenario.steps:
            name = step.get('name', 'Unnamed Step')
            cond = step.get('condition', 'OR')
            imgs = ','.join([
                f"{img.get('name', 'img')} ({img.get('sensitivity', 0.9):.2f})"
                for img in step.get('images', [])
            ])
            acts = ','.join([a.get('type', 'action') for a in step.get('actions', [])])
            self.steps_list.addItem(f"{name} [{cond}] imgs: {imgs} actions: {acts}")

    def refresh_window_list(self, keep_selection=False):
        """
        Populate the window selection dropdown with all open windows and 'Entire Screen'.
        If keep_selection is True, try to keep the current selection.
        """
        self.window_combo.blockSignals(True)
        current = self.window_combo.currentText() if keep_selection else None
        self.window_combo.clear()
        self.window_combo.addItem('Entire Screen')
        try:
            windows = gw.getAllTitles()
            for title in windows:
                if title.strip():
                    self.window_combo.addItem(title)
        except Exception as e:
            logger.error(f'Error listing windows: {e}')
        if keep_selection and current:
            idx = self.window_combo.findText(current)
            if idx != -1:
                self.window_combo.setCurrentIndex(idx)
        self.window_combo.blockSignals(False)

    def save_selected_window(self):
        """
        Save the currently selected window to the scenario's config file.
        """
        if self.current_scenario:
            selected_window = self.window_combo.currentText()
            scenario_dir = self.current_scenario.get_scenario_dir()
            config_path = os.path.join(scenario_dir, 'window_config.json')
            try:
                with open(config_path, 'w') as f:
                    json.dump({'selected_window': selected_window}, f)
            except Exception as e:
                logger.error(f'Failed to save window selection: {e}')

    def load_selected_window_from_config(self):
        """
        Load the selected window from the scenario's config file and update the dropdown.
        """
        if self.current_scenario:
            scenario_dir = self.current_scenario.get_scenario_dir()
            config_path = os.path.join(scenario_dir, 'window_config.json')
            
            # Default to 'Entire Screen'
            win_name = 'Entire Screen'
            
            if os.path.exists(config_path):
                try:
                    with open(config_path, 'r') as f:
                        data = json.load(f)
                        win_name = data.get('selected_window', 'Entire Screen')
                except Exception as e:
                    logger.error(f'Failed to load window selection: {e}')

            # Use a single shot timer to set the index after the event loop has processed
            def set_window():
                self.refresh_window_list() # Refresh list first
                idx = self.window_combo.findText(win_name)
                if idx != -1:
                    self.window_combo.setCurrentIndex(idx)
                    logger.info(f"UI: Set window to '{win_name}' from config.")
                else:
                    self.window_combo.setCurrentIndex(0)
                    logger.warning(f"UI: Saved window '{win_name}' not found. Defaulting to 'Entire Screen'.")

            QtCore.QTimer.singleShot(100, set_window)
# StepDialog for creating/editing a step
class StepDialog(QtWidgets.QDialog):
    """
    Dialog for creating or editing a step in a scenario.
    Allows editing images, actions, and step properties.
    """
    def __init__(self, parent=None, step=None):
        """
        Initialize the StepDialog for creating or editing a step.
        """
        super().__init__(parent)
        self.setWindowTitle('Step Editor')
        self.resize(600, 400)
        self.images = [] if step is None else step.get('images', [])
        self.actions = [] if step is None else step.get('actions', [])
        self.name_edit = QtWidgets.QLineEdit(step['name'] if step and 'name' in step else '')
        self.cond_combo = QtWidgets.QComboBox()
        self.cond_combo.addItems(['OR', 'AND'])
        if step and 'condition' in step:
            self.cond_combo.setCurrentText(step['condition'])
        # Images Group
        images_group = QtWidgets.QGroupBox('Images')
        images_group_layout = QtWidgets.QVBoxLayout()
        images_row_layout = QtWidgets.QHBoxLayout()
        img_list_layout = QtWidgets.QVBoxLayout()
        self.img_list = QtWidgets.QListWidget()
        self.img_list.currentItemChanged.connect(self.update_image_preview)
        for img in self.images:
            self.img_list.addItem(img.get('name', 'img'))
        img_list_layout.addWidget(self.img_list)
        img_list_layout.addWidget(QtWidgets.QLabel('Sensitivity:'))
        self.sensitivity_spin = QtWidgets.QDoubleSpinBox()
        self.sensitivity_spin.setRange(0.5, 1.0)
        self.sensitivity_spin.setSingleStep(0.01)
        self.sensitivity_spin.setDecimals(2)
        self.sensitivity_spin.setValue(0.9)
        self.sensitivity_spin.setToolTip('Sensitivity for selected image (higher = stricter match)')
        self.img_list.currentRowChanged.connect(self.update_sensitivity_spin)
        self.sensitivity_spin.valueChanged.connect(self.save_sensitivity_for_image)
        img_list_layout.addWidget(self.sensitivity_spin)
        images_row_layout.addLayout(img_list_layout)
        self.img_preview = QtWidgets.QLabel('Image Preview')
        self.img_preview.setAlignment(QtCore.Qt.AlignmentFlag.AlignCenter)
        self.img_preview.setFixedSize(200, 200)
        images_row_layout.addWidget(self.img_preview)
        images_group_layout.addLayout(images_row_layout)
        # Image buttons horizontal, spanning full width
        img_btns_layout = QtWidgets.QHBoxLayout()
        self.btn_add_img = QtWidgets.QPushButton('Add Image')
        self.btn_add_img.clicked.connect(self.add_image_to_step)
        self.btn_del_img = QtWidgets.QPushButton('Delete Image')
        self.btn_del_img.clicked.connect(self.delete_image)
        self.btn_rename_img = QtWidgets.QPushButton("Rename Image")
        self.btn_rename_img.clicked.connect(self.rename_image)
        self.btn_retake_img = QtWidgets.QPushButton("Retake Screenshot")
        self.btn_retake_img.clicked.connect(self.retake_screenshot)
        img_btns_layout.addWidget(self.btn_add_img)
        img_btns_layout.addWidget(self.btn_del_img)
        img_btns_layout.addWidget(self.btn_rename_img)
        img_btns_layout.addWidget(self.btn_retake_img)
        images_group_layout.addLayout(img_btns_layout)
        images_group.setLayout(images_group_layout)

        # Actions Group
        actions_group = QtWidgets.QGroupBox('Actions')
        actions_group_layout = QtWidgets.QVBoxLayout()
        self.act_list = QtWidgets.QListWidget()
        for act in self.actions:
            self.act_list.addItem(act.get('type', 'action'))
        actions_group_layout.addWidget(self.act_list)
        act_btns_layout = QtWidgets.QHBoxLayout()
        self.btn_add_act = QtWidgets.QPushButton('Add Action')
        self.btn_add_act.clicked.connect(self.add_action)
        self.btn_del_act = QtWidgets.QPushButton('Delete Action')
        self.btn_del_act.clicked.connect(self.delete_action)
        self.btn_move_up_act = QtWidgets.QPushButton('Move Up')
        self.btn_move_up_act.clicked.connect(self.move_action_up)
        self.btn_move_down_act = QtWidgets.QPushButton('Move Down')
        self.btn_move_down_act.clicked.connect(self.move_action_down)
        act_btns_layout.addWidget(self.btn_add_act)
        act_btns_layout.addWidget(self.btn_del_act)
        act_btns_layout.addWidget(self.btn_move_up_act)
        act_btns_layout.addWidget(self.btn_move_down_act)
        actions_group_layout.addLayout(act_btns_layout)
        actions_group.setLayout(actions_group_layout)

        # Main Layout
        layout = QtWidgets.QGridLayout()
        layout.addWidget(QtWidgets.QLabel('Step Name:'), 0, 0)
        layout.addWidget(self.name_edit, 0, 1, 1, 2)
        layout.addWidget(QtWidgets.QLabel('Condition:'), 1, 0)
        layout.addWidget(self.cond_combo, 1, 1, 1, 2)
        layout.addWidget(images_group, 2, 0, 1, 3)
        layout.addWidget(actions_group, 3, 0, 1, 3)
        btns = QtWidgets.QDialogButtonBox(QtWidgets.QDialogButtonBox.StandardButton.Ok | QtWidgets.QDialogButtonBox.StandardButton.Cancel)
        btns.accepted.connect(self.accept)
        btns.rejected.connect(self.reject)
        layout.addWidget(btns, 4, 1, 1, 2)
        self.setLayout(layout)

    def move_action_up(self):
        """
        Move the selected action up in the action list.
        """
        idx = self.act_list.currentRow()
        if idx > 0:
            self.actions[idx-1], self.actions[idx] = self.actions[idx], self.actions[idx-1]
            item = self.act_list.takeItem(idx)
            self.act_list.insertItem(idx-1, item)
            self.act_list.setCurrentRow(idx-1)

    def move_action_down(self):
        """
        Move the selected action down in the action list.
        """
        idx = self.act_list.currentRow()
        if idx < len(self.actions) - 1 and idx >= 0:
            self.actions[idx+1], self.actions[idx] = self.actions[idx], self.actions[idx+1]
            item = self.act_list.takeItem(idx)
            self.act_list.insertItem(idx+1, item)
            self.act_list.setCurrentRow(idx+1)

    class NameImageDialog(QtWidgets.QDialog):
        def __init__(self, pixmap, parent=None):
            """
            Initialize the dialog for naming a new image.
            """
            super().__init__(parent)
            self.setWindowTitle("Name Your Image")
            
            # Layouts
            layout = QtWidgets.QVBoxLayout(self)
            form_layout = QtWidgets.QFormLayout()
            
            # Image Preview
            self.preview_label = QtWidgets.QLabel()
            self.preview_label.setFixedSize(300, 300)
            self.preview_label.setPixmap(pixmap.scaled(300, 300, QtCore.Qt.AspectRatioMode.KeepAspectRatio, QtCore.Qt.TransformationMode.SmoothTransformation))
            layout.addWidget(self.preview_label, alignment=QtCore.Qt.AlignmentFlag.AlignCenter)
            
            # Name Input
            self.name_edit = QtWidgets.QLineEdit()
            form_layout.addRow("Image Name:", self.name_edit)
            layout.addLayout(form_layout)
            
            # Dialog Buttons
            button_box = QtWidgets.QDialogButtonBox(QtWidgets.QDialogButtonBox.StandardButton.Ok | QtWidgets.QDialogButtonBox.StandardButton.Cancel)
            button_box.accepted.connect(self.accept)
            button_box.rejected.connect(self.reject)
            layout.addWidget(button_box)

        def get_name(self):
            """
            Return the entered image name.
            """
            return self.name_edit.text()

    def add_image_to_step(self):
        """
        Add a new image to the step by taking a screenshot and letting the user select a region.
        Optimized for memory efficiency.
        """
        main_window = self.parent()
        step_name = self.name_edit.text()
        if not step_name:
            QtWidgets.QMessageBox.warning(self, "Step Name Required", "Please enter a name for the step before adding an image.")
            return

        main_window.hide()
        self.hide()
        time.sleep(0.3)

        cropped_pil_image = None
        rect_coords = None
        try:
            rect_coords = take_screenshot_with_tkinter()
            if rect_coords:
                # Use more memory-efficient screenshot capture
                bbox = (
                    rect_coords['x'], 
                    rect_coords['y'], 
                    rect_coords['x'] + rect_coords['width'], 
                    rect_coords['y'] + rect_coords['height']
                )
                cropped_pil_image = ImageGrab.grab(bbox=bbox)
            logger.debug("StepDialog.add_image_to_step: Screenshot selection finished.")
        except Exception as e:
            logger.error(f"StepDialog.add_image_to_step: Screenshot failed: {e}")
            QtWidgets.QMessageBox.critical(self, 'Error', f'Failed to take screenshot: {e}')
        finally:
            self.show()
            main_window.show()

        if cropped_pil_image and rect_coords:
            try:
                # Convert PIL to QPixmap more efficiently
                img_np = np.array(cropped_pil_image.convert('RGB'))
                height, width, channel = img_np.shape
                bytes_per_line = 3 * width
                qimage = QtGui.QImage(img_np.data, width, height, bytes_per_line, QtGui.QImage.Format.Format_RGB888)
                cropped_pixmap = QtGui.QPixmap.fromImage(qimage)

                name_dialog = self.NameImageDialog(cropped_pixmap, self)
                if name_dialog.exec():
                    name = name_dialog.get_name()
                    if name:
                        scenario_dir = main_window.current_scenario.get_scenario_dir()
                        step_images_dir = os.path.join(scenario_dir, "steps", step_name)
                        if not os.path.exists(step_images_dir):
                            os.makedirs(step_images_dir)

                        safe_name = "".join(c for c in name if c.isalnum() or c in (' ', '_')).rstrip()
                        path = os.path.join(step_images_dir, f'{safe_name}.png')

                        # Save with PNG optimization
                        if cropped_pixmap.save(path, "PNG", quality=90):
                            logger.info(f"Screenshot saved to {path}")
                            img_obj = {
                                'path': path, 
                                'region': [rect_coords['x'], rect_coords['y'], rect_coords['width'], rect_coords['height']], 
                                'name': name, 
                                'sensitivity': 0.9
                            }
                            self.images.append(img_obj)
                            self.img_list.addItem(name)
                        else:
                            logger.error(f"Failed to save screenshot to {path}")
                            QtWidgets.QMessageBox.critical(self, 'Error', 'Failed to save the screenshot file.')
                
                # Clean up memory
                del img_np, qimage, cropped_pixmap
                
            finally:
                # Ensure PIL image is cleaned up
                if cropped_pil_image:
                    cropped_pil_image.close()
                    del cropped_pil_image
                gc.collect()

    def update_image_preview(self, current, previous):
        """
        Update the image preview label when a new image is selected.
        Optimized to prevent memory leaks from large pixmaps.
        """
        # Clear previous pixmap to free memory
        if previous:
            self.img_preview.clear()
        
        if current:
            try:
                idx = self.img_list.row(current)
                if idx < 0 or idx >= len(self.images):
                    self.img_preview.setText('Image Preview')
                    return
                
                path = self.images[idx]['path']
                if not os.path.exists(path):
                    self.img_preview.setText('Image Not Found')
                    return
                
                # Load and scale image efficiently
                pixmap = QtGui.QPixmap(path)
                if pixmap.isNull():
                    self.img_preview.setText('Invalid Image')
                    return
                
                # Scale with memory optimization
                preview_size = self.img_preview.size()
                if pixmap.width() > preview_size.width() or pixmap.height() > preview_size.height():
                    scaled_pixmap = pixmap.scaled(
                        preview_size, 
                        QtCore.Qt.AspectRatioMode.KeepAspectRatio, 
                        QtCore.Qt.TransformationMode.SmoothTransformation
                    )
                    self.img_preview.setPixmap(scaled_pixmap)
                    # Clear original large pixmap
                    del pixmap
                else:
                    self.img_preview.setPixmap(pixmap)
                    
            except Exception as e:
                logger.error(f"Error updating image preview: {e}")
                self.img_preview.setText('Preview Error')
        else:
            self.img_preview.setText('Image Preview')

    def update_sensitivity_spin(self, idx):
        """
        Update the sensitivity spinbox for the selected image.
        """
        if idx < 0 or idx >= len(self.images):
            self.sensitivity_spin.setValue(0.9)
            self.sensitivity_spin.setEnabled(False)
            return
        self.sensitivity_spin.setEnabled(True)
        img = self.images[idx]
        self.sensitivity_spin.setValue(img.get('sensitivity', 0.9))

    def save_sensitivity_for_image(self, value):
        """
        Save the sensitivity value for the currently selected image.
        """
        idx = self.img_list.currentRow()
        if idx < 0 or idx >= len(self.images):
            return
        self.images[idx]['sensitivity'] = value

    def delete_image(self):
        """
        Delete the currently selected image from the step.
        """
        idx = self.img_list.currentRow()
        if idx >= 0:
            self.images.pop(idx)
            self.img_list.takeItem(idx)

    def rename_image(self):
        """
        Open a dialog to rename the currently selected image.
        """
        idx = self.img_list.currentRow()
        if idx < 0:
            return

        img_obj = self.images[idx]
        old_name = img_obj.get('name', '')
        new_name, ok = QtWidgets.QInputDialog.getText(self, "Rename Image", "New name:", text=old_name)

        if ok and new_name and new_name != old_name:
            img_obj['name'] = new_name
            self.img_list.item(idx).setText(new_name)

    def retake_screenshot(self):
        """
        Retake the screenshot for the currently selected image.
        """
        idx = self.img_list.currentRow()
        if idx < 0:
            return

        main_window = self.parent()
        main_window.hide()
        self.hide()
        time.sleep(0.3)

        cropped_pil_image = None
        rect_coords = None
        try:
            rect_coords = take_screenshot_with_tkinter()
            if rect_coords:
                cropped_pil_image = ImageGrab.grab(bbox=(
                    rect_coords['x'], 
                    rect_coords['y'], 
                    rect_coords['x'] + rect_coords['width'], 
                    rect_coords['y'] + rect_coords['height']
                ))
            logger.debug("StepDialog.retake_screenshot: Screenshot selection finished.")
        except Exception as e:
            logger.error(f"StepDialog.retake_screenshot: Screenshot failed: {e}")
            QtWidgets.QMessageBox.critical(self, 'Error', f'Failed to take screenshot: {e}')
        finally:
            self.show()
            main_window.show()

        if cropped_pil_image and rect_coords:
            img_np = np.array(cropped_pil_image.convert('RGB'))
            height, width, channel = img_np.shape
            bytes_per_line = 3 * width
            qimage = QtGui.QImage(img_np.data, width, height, bytes_per_line, QtGui.QImage.Format.Format_RGB888)
            cropped_pixmap = QtGui.QPixmap.fromImage(qimage)

            img_obj = self.images[idx]
            path = img_obj['path']

            if cropped_pixmap.save(path, "PNG"):
                logger.info(f"Screenshot retaken and saved to {path}")
                img_obj['region'] = [rect_coords['x'], rect_coords['y'], rect_coords['width'], rect_coords['height']]
                self.update_image_preview(self.img_list.currentItem(), None)
            else:
                logger.error(f"Failed to save retaken screenshot to {path}")
                QtWidgets.QMessageBox.critical(self, 'Error', 'Failed to save the retaken screenshot file.')

    def add_action(self):
        dlg = StepActionDialog(self)
        if dlg.exec():
            act = dlg.get_action()
            self.actions.append(act)
            self.act_list.addItem(act.get('type', 'action'))

    def delete_action(self):
        """
        Delete the currently selected action from the step.
        """
        idx = self.act_list.currentRow()
        if idx >= 0:
            self.actions.pop(idx)
            self.act_list.takeItem(idx)

    def get_step(self):
        """
        Return the step as a dictionary for saving.
        """
        return {
            'name': self.name_edit.text(),
            'condition': self.cond_combo.currentText(),
            'images': self.images,
            'actions': self.actions
        }

# StepActionDialog for adding actions to a step
class StepActionDialog(QtWidgets.QDialog):
    """
    Dialog for adding or editing an action for a step.
    Supports click, key, scroll, and delay actions.
    """
    def __init__(self, parent=None):
        """
        Initialize the StepActionDialog for adding or editing a step action.
        """
        super().__init__(parent)
        self.setWindowTitle('Add Action')
        self.type_combo = QtWidgets.QComboBox()
        self.type_combo.addItems(['click', 'key', 'scroll', 'delay'])
        # Click options
        self.click_btn_combo = QtWidgets.QComboBox()
        self.click_btn_combo.addItems(['left', 'middle', 'right'])
        self.click_pos_combo = QtWidgets.QComboBox()
        self.click_pos_combo.addItems(['center', 'relative', 'absolute'])
        self.rel_x_spin = QtWidgets.QSpinBox()
        self.rel_x_spin.setRange(-10000, 10000)
        self.rel_y_spin = QtWidgets.QSpinBox()
        self.rel_y_spin.setRange(-10000, 10000)
        self.abs_x_spin = QtWidgets.QSpinBox()
        self.abs_x_spin.setRange(0, 10000)
        self.abs_y_spin = QtWidgets.QSpinBox()
        self.abs_y_spin.setRange(0, 10000)
        self.pick_abs_btn = QtWidgets.QPushButton('Pick Position')
        self.pick_abs_btn.clicked.connect(self.pick_absolute_position)
        # Scroll options
        self.scroll_dir_combo = QtWidgets.QComboBox()
        self.scroll_dir_combo.addItems(['up', 'down'])
        self.scroll_amt_spin = QtWidgets.QSpinBox()
        self.scroll_amt_spin.setRange(1, 10000)
        self.scroll_amt_spin.setValue(100)
        # Key options
        self.key_label = QtWidgets.QLabel('(press a key)')
        self.key_value = None
        self.key_capture_btn = QtWidgets.QPushButton('Set Key')
        self.key_capture_btn.clicked.connect(self.capture_key)
        # Delay options
        self.delay_spin = QtWidgets.QDoubleSpinBox()
        self.delay_spin.setRange(0.1, 60.0)
        self.delay_spin.setSingleStep(0.1)
        self.delay_spin.setValue(1.0)
        # Layout
        self.form = QtWidgets.QFormLayout()
        self.form.addRow('Type:', self.type_combo)
        # Click group
        self.click_group = QtWidgets.QGroupBox('Click Options')
        click_layout = QtWidgets.QFormLayout()
        click_layout.addRow('Button:', self.click_btn_combo)
        click_layout.addRow('Position Mode:', self.click_pos_combo)
        click_layout.addRow('Relative X:', self.rel_x_spin)
        click_layout.addRow('Relative Y:', self.rel_y_spin)
        click_layout.addRow('Absolute X:', self.abs_x_spin)
        click_layout.addRow('Absolute Y:', self.abs_y_spin)
        click_layout.addRow(self.pick_abs_btn)
        self.click_group.setLayout(click_layout)
        self.form.addRow(self.click_group)
        # Scroll group
        self.scroll_group = QtWidgets.QGroupBox('Scroll Options')
        scroll_layout = QtWidgets.QFormLayout()
        scroll_layout.addRow('Direction:', self.scroll_dir_combo)
        scroll_layout.addRow('Amount:', self.scroll_amt_spin)
        self.scroll_group.setLayout(scroll_layout)
        self.form.addRow(self.scroll_group)
        # Key group
        self.key_group = QtWidgets.QGroupBox('Key Options')
        key_layout = QtWidgets.QHBoxLayout()
        key_layout.addWidget(self.key_label)
        key_layout.addWidget(self.key_capture_btn)
        self.key_group.setLayout(key_layout)
        self.form.addRow(self.key_group)
        # Delay group
        self.delay_group = QtWidgets.QGroupBox('Delay Options')
        delay_layout = QtWidgets.QFormLayout()
        delay_layout.addRow('Duration (s):', self.delay_spin)
        self.delay_group.setLayout(delay_layout)
        self.form.addRow(self.delay_group)
        # Buttons
        btns = QtWidgets.QDialogButtonBox(QtWidgets.QDialogButtonBox.StandardButton.Ok | QtWidgets.QDialogButtonBox.StandardButton.Cancel)
        btns.accepted.connect(self.accept)
        btns.rejected.connect(self.reject)
        self.form.addWidget(btns)
        self.setLayout(self.form)
        self.type_combo.currentTextChanged.connect(lambda: self.update_fields())
        self.click_pos_combo.currentTextChanged.connect(lambda: self.update_fields())
        self.update_fields()

    def update_fields(self):
        """
        Update the visibility of fields based on the selected action type.
        """
        t = self.type_combo.currentText()
        self.click_group.setVisible(t == 'click')
        self.scroll_group.setVisible(t == 'scroll')
        self.key_group.setVisible(t == 'key')
        self.delay_group.setVisible(t == 'delay')
        # Click position fields
        if t == 'click':
            self.click_group.setVisible(True)
            pos_mode = self.click_pos_combo.currentText()
            # Find the layout rows for rel_x, rel_y, abs_x, abs_y
            for i in range(self.click_group.layout().rowCount()):
                label = self.click_group.layout().itemAt(i, QtWidgets.QFormLayout.ItemRole.LabelRole)
                field = self.click_group.layout().itemAt(i, QtWidgets.QFormLayout.ItemRole.FieldRole)
                if label and field:
                    label_text = label.widget().text() if hasattr(label.widget(), 'text') else ''
                    if label_text == 'Relative X:' or label_text == 'Relative Y:':
                        show = pos_mode == 'relative'
                        label.widget().setVisible(show)
                        field.widget().setVisible(show)
                    elif label_text == 'Absolute X:' or label_text == 'Absolute Y:':
                        show = pos_mode == 'absolute'
                        label.widget().setVisible(show)
                        field.widget().setVisible(show)
            self.pick_abs_btn.setVisible(pos_mode == 'absolute')
        else:
            self.click_group.setVisible(False)

    def pick_absolute_position(self):
        """
        Open an overlay to pick an absolute position on the screen.
        """
        # Overlay to pick a position on screen
        class PosOverlay(QtWidgets.QWidget):
            pos_picked = QtCore.pyqtSignal(int, int)
            def __init__(self):
                super().__init__()
                self.setWindowFlags(QtCore.Qt.WindowType.FramelessWindowHint | QtCore.Qt.WindowType.WindowStaysOnTopHint)
                self.setWindowState(QtCore.Qt.WindowState.WindowFullScreen)
                self.setAttribute(QtCore.Qt.WidgetAttribute.WA_TranslucentBackground)
                self.setCursor(QtGui.QCursor(QtCore.Qt.CursorShape.CrossCursor))
            def mousePressEvent(self, event):
                self.pos_picked.emit(event.globalX(), event.globalY())
                self.close()
            def paintEvent(self, event):
                pass
        overlay = PosOverlay()
        overlay.pos_picked.connect(self.set_absolute_position)
        overlay.showFullScreen()

    def set_absolute_position(self, x, y):
        """
        Set the absolute position fields to the picked coordinates.
        """
        self.abs_x_spin.setValue(x)
        self.abs_y_spin.setValue(y)

    def capture_key(self):
        """
        Open a dialog to capture a key press from the user.
        """
        # Open a modal dialog to capture a key press
        dlg = QtWidgets.QDialog(self)
        dlg.setWindowTitle('Press a key')
        label = QtWidgets.QLabel('Press any key...', dlg)
        layout = QtWidgets.QVBoxLayout()
        layout.addWidget(label)
        dlg.setLayout(layout)
        key_captured = {}
        def on_key(event):
            if event.type() == QtCore.QEvent.Type.KeyPress:
                key_captured['key'] = event.key()
                key_captured['text'] = event.text()
                dlg.accept()
                return True
            return False
        dlg.installEventFilter(self)
        orig_event = dlg.event
        def event_filter(obj, event):
            if event.type() == QtCore.QEvent.Type.KeyPress:
                key = event.key()
                text = event.text()
                self.key_value = text if text else str(key)
                self.key_label.setText(f'Key: {self.key_value}')
                dlg.accept()
                return True
            return orig_event(event)
        dlg.event = event_filter
        dlg.exec()

    def get_action(self):
        """
        Return the action as a dictionary for saving.
        """
        t = self.type_combo.currentText()
        params = {}
        if t == 'click':
            params['button'] = self.click_btn_combo.currentText()
            pos_mode = self.click_pos_combo.currentText()
            params['pos_type'] = pos_mode
            if pos_mode == 'relative':
                params['rel_x'] = self.rel_x_spin.value()
                params['rel_y'] = self.rel_y_spin.value()
            elif pos_mode == 'absolute':
                params['abs_x'] = self.abs_x_spin.value()
                params['abs_y'] = self.abs_y_spin.value()
        elif t == 'scroll':
            params['direction'] = self.scroll_dir_combo.currentText()
            params['amount'] = self.scroll_amt_spin.value()
        elif t == 'key':
            params['key'] = self.key_value if self.key_value else ''
        elif t == 'delay':
            params['duration'] = self.delay_spin.value()
        return {'type': t, 'params': params}

    
    def perform_step_action(self, action, loc, shape):
        """
        Perform the given step action at the specified location.
        """
        act_type = action['type']
        params = action['params']
        logger.debug(f'Performing step action: {act_type}, params={params}, loc={loc}, shape={shape}')
        try:
            if act_type == 'click':
                pos_type = params.get('pos_type', 'center')
                if pos_type == 'center':
                    x = loc[0] + shape[1] // 2
                    y = loc[1] + shape[0] // 2
                elif pos_type == 'relative':
                    x = loc[0] + params.get('rel_x', 0)
                    y = loc[1] + params.get('rel_y', 0)
                elif pos_type == 'absolute':
                    x = params.get('abs_x', 0)
                    y = params.get('abs_y', 0)
                else:
                    x, y = loc[0], loc[1]
                button = params.get('button', 'left')
                logger.info(f'Clicking at ({x}, {y}) with button {button}')
                pyautogui.click(x, y, button=button)
            elif act_type == 'key':
                key = params.get('key', 'enter')
                logger.info(f'Pressing key: {key}')
                pyautogui.press(key)
            elif act_type == 'scroll':
                amount = params.get('amount', 0)
                direction = params.get('direction', 'up')
                logger.info(f'Scrolling: {amount} direction: {direction}')
                if direction == 'up':
                    pyautogui.scroll(amount)
                else:
                    pyautogui.scroll(-abs(amount))
            elif act_type == 'delay':
                duration = params.get('duration', 1)
                logger.info(f'Delay for {duration} seconds')
                time.sleep(duration)
        except Exception as e:
            logger.error(f'Error performing step action {act_type}: {e}')

class ActionDialog(QtWidgets.QDialog):
    """
    Dialog for adding a generic action to a scenario (legacy/unused in main flow).
    """
    def __init__(self, parent=None, images=None):
        """
        Initialize the ActionDialog for adding a generic action.
        """
        super().__init__(parent)
        self.setWindowTitle('Add Action')
        self.type_combo = QtWidgets.QComboBox()
        self.type_combo.addItems(['click', 'key', 'scroll'])
        self.param_edit = QtWidgets.QLineEdit()
        self.delay_before = QtWidgets.QSpinBox()
        self.delay_after = QtWidgets.QSpinBox()
        self.delay_before.setMaximum(10000)
        self.delay_after.setMaximum(10000)
        self.images = images or []
        self.image_checks = []
        self.logic_combo = QtWidgets.QComboBox()
        self.logic_combo.addItems(['OR', 'AND'])
        layout = QtWidgets.QFormLayout()
        layout.addRow('Type:', self.type_combo)
        layout.addRow('Params (JSON):', self.param_edit)
        layout.addRow('Delay Before (s):', self.delay_before)
        layout.addRow('Delay After (s):', self.delay_after)
        layout.addRow('Image Logic:', self.logic_combo)
        img_group = QtWidgets.QGroupBox('Trigger Images')
        img_layout = QtWidgets.QVBoxLayout()
        for img in self.images:
            cb = QtWidgets.QCheckBox(img.get('name', os.path.basename(img['path'])))
            img_layout.addWidget(cb)
            self.image_checks.append(cb)
        img_group.setLayout(img_layout)
        layout.addRow(img_group)
        btns = QtWidgets.QDialogButtonBox(QtWidgets.QDialogButtonBox.StandardButton.Ok | QtWidgets.QDialogButtonBox.StandardButton.Cancel)
        btns.accepted.connect(self.accept)
        btns.rejected.connect(self.reject)
        layout.addWidget(btns)
        self.setLayout(layout)

    def get_action(self):
        """
        Return the action as a dictionary for saving.
        """
        try:
            params = json.loads(self.param_edit.text())
        except Exception:
            params = {}
        params['delay_before'] = self.delay_before.value()
        params['delay_after'] = self.delay_after.value()
        image_refs = [img.get('name', os.path.basename(img['path'])) for img, cb in zip(self.images, self.image_checks) if cb.isChecked()]
        logic = self.logic_combo.currentText()
        return {'type': self.type_combo.currentText(), 'params': params, 'image_refs': image_refs, 'logic': logic}

if __name__ == '__main__':
    app = QtWidgets.QApplication(sys.argv)
    win = MainWindow()
    win.show()
    sys.exit(app.exec())