I've made a simple click bot to automatically play an android game on my Windows PC. It currently identifies when certain things change on the screen then moves the mouse and clicks the correct button.
Currently I am using the following win32api functions to achieve this:
Win32api.SetCursorPos(Position)
Win32api.mouse_event(winn32con.MOUSEEVENTF_LEFTDOWN,0,0)
Win32api.mouse_event(winn32con.MOUSEEVENTF_LEFTUP,0,0)
These work great however when I use the bot it takes over my computer's mouse and I basically have to let it run. Is there anyway I could simulate a click without it actually using my mouse, or is there a way I could isolate the bot on one of my other screens and be able to work freely on the other?
There is a lib specific for deal with user interaction components and periferics: pyautogui
Here, is a short and easy to try documentation for performing/simulating mouse click DOWN & UP
https://pyautogui.readthedocs.org/en/latest/mouse.html#the-mousedown-and-mouseup-functions
Is there a way to turn off the notificationCenter in iOS game using cocos2d-x? Suppose in my game, there are objects which have to be dragged from the top. The notificationCenter interrupts in between. Please anyone here tell me how to turn off the notification center when the game runs.
No, the Notification Center is a system level feature. And iOS is not exposed any API to turn it off in some applications.
If by notificationCenter your mean CCNotificationCenter then you can call CCNotificationCenter::purgeNotificationCenter to remove all the observers or CCNotificationCenter::removeAllObservers(target) to stop notifications for a specific observer.
Notification center could not be disabled in the app.
According to How to disable the "drag down to view Notifications" feature?, the only way to stop the Notification Center from immediately appearing is to hide your app's status bar, and even then the 'tab' of the NC will stay appear.
Like Multitasking Gestures in iOS5, you cannot stop this behavior and must instead rethink how your apps work to suit Apple's changes.
if ([self respondsToSelector:#selector(setNeedsStatusBarAppearanceUpdate)]) {
// iOS 7
[self performSelector:#selector(setNeedsStatusBarAppearanceUpdate)];
} else {
// iOS 6
[[UIApplication sharedApplication] setStatusBarHidden:YES
withAnimation:UIStatusBarAnimationSlide];
}
- (BOOL)prefersStatusBarHidden {
return YES;
}
I have a simple game, breakout clone.
There is gameplay layer, menu layer. I can call menu during play just fine, can use next level button ( it makes new gameplay layer with next level and transitions to its scene). But what i want is at the end of level, when I killed all blocks I want that menu to pop up so player could click next level. But the provlem is that i store blocks number in gameplay layer, so when they reach zero i get certain property turned to true. But when i try to read that property from menu when level ends, i can't, because my gameplay layer is already deallocated since i transited to menu.
Tldr; how to change level with button?
I think your question is about Game Architecture. There are lots of trade-offs regarding how you choose to architect your game. This is something that I have thought a lot about as I have created Cocos2d games.
The comment by Shailesh_ios will work. But if you want to create a more complex game you should consider designing the game in a way that neatly organizes the different modules and components in your game so that they are easy to use and support. As your game gets more complex it can become very messy if you add loads of functionality to the AppDelgate. To stay organized and keep your sanity when updating your code, I suggest creating a centralized game manager service in your game.
Create a Centralized GameManager Service
--For all centralized services that your game will need, one thing that has worked good for me is to create a game manager class that is a CCNode or NSObject subclass. This class is made into a singleton. I then create separate classes for things like GameCenter, OpenFeint, In-App Purchase, player preferences, local scores and achievements, physics world manager, game audio, game state, etc.
Add class instances as modules to the GameManager
--These classes are all then owned by the GameManager singleton class and can provide services to the game components that need them. Since these classes generally use very little memory there is no need to constantly be creating and destroying them each time a scene changes, which just wastes CPU cycles and battery life.
A note of caution
--Be careful though, because it is easy to abuse the global singleton GameManager class by adding things to it that don't belong in it. Any centralized service that may be requested from various game components could be considered to be added to the GameManager. By adding to the GameManager, I mean coding each module in its own class and then have it be created and owned by the GameManager.
Examples--
For example, your custom local scores and achievements class could be setup to store a players scores, persist them to a file or NSUserDefaults, and make that information available to any game component that needs it.
The custom GameCenter or OpenFeint class would take care of authenticating the player and then as a service provide submitting scores/achievements, downloading scores/achievements, presenting the built in GameCenter/OpenFeint UI for leaderboards and achievements, etc.
This way you can focus on designing your game and have the basic services a game needs available from anywhere within your game.
How to know if a mouse click is simulated or not? When mouse click send by a program or real mouse device
... I'm programming a system detection for a game to avoid bots, autoclicks,etc that only accept legit mouse clicks
This depends a bit on the kind of application you are writing, but if you can, I would watch the cursor movement, not the clicks.
Human mouse movement has non-uniform speeds, reaction times, imprecisions (clicks on different coordinates of your buttons, etc...).
Also, you can defend a gui against bots by randomly requiring an interaction that is hard to script. For example: If scripts depend upon buttons being always in the same position, I would make sure that, while trying to remain intuitive, the dialog should pop up in slightly different positions every time.
Otherwise: There is no way to detect if the mouse is a real one or a really well simulated one. The Windows HID/MacOS/Linux driver layer abstracts away the distinction between Mice, TrackPens, TrackBalls, draw-pads, touch screens... and of course script-mice...
Although the blog post itself is about a different issue, I refer you to Raymond Chen's excellent Old New Thing. In this specific blog post he talks about the validity of message parameters going into an application, but also makes the point that:
There's no point discussing the possibility that the sender of the message is playing tricks and lying to you because (1) your program should just go along with the ruse and respond to fake menu messages as if they were real menu messages, because (2) there's no way to tell that you're being lied to anyway. To detect lying, you'd have to be able to read into the mindset of the programmer who sent you the message.
Essentially the argument is that you should respond to mouse clicks as mouse clicks, regardless of how those clicks were generated.
Is mouse keys simulated mouse input or legit? The point of simulating mouse input is to make them look exactly like real mouse input. If the simulation is doing its job, then your job is impossible. Sorry, that's the blessing & curse of software for you. Here are some more imperfect ideas:
Use GetKeyboardState and verify that the button states are correct. If the message faker is using PostMessage, they will likely not be setting keyboard state and this would indicate fakery.
If you are targeting known applications that are doing the input simulation, detect them and complain. This is not perfect at all for many reasons.
Fuzzy logic, as many other people have suggested.
You need to be creative and figure out the difference between a simulated event and a real one to you, as there is no generalized answer.
It can't be done (reliably (with software alone anyway))
I've used WIN32API calls to read pixels/manipulate the mouse/send keystrokes to automate large portions of video games and other repetitive tasks. You could write a lot of code to analyze the input, but equally smart developers are just going to modify their code to match.
When I first try to automate a mouse click, that's all I'll do. Send a mouse click. And most of the time it works. You might have code that tracks the mouse movement and the entire stack of mouse events that would fire along with a legitimate click and say, 'That wasn't real - we ignore it' but nothing stops the developer from also implementing mouse movements.
The mouse events are more complex than keypresses; but it's essentially the same idea. If you write code that monitors the time between keypresses and determine that I'm sending the '2' key to your application in EXACTLY 250ms intervals, you might decide I'm a bot. But, all I'll do is modify my code to send the keystroke in 250ms + a random value between -25 and 25 ms.
It's a never-ending game of cat and mouse. The best solution is to make tasks non-trivial so simple forms of automation aren't applicable.
The question is a bit thin on details.
Events can be sent directly to controls without moving the mouse so find out where the mouse is when you get the click event and see if it's on the control. Keyboard input requires control focus, so check that too.
For situations where the mouse moves, you won't be able to tell if the mouse movement is recorded and played back. If its scripted then perhaps you could monitor the mouse behavior in the parent panel(s) of the control and use those events and movements to ascertain whether it is real or not. An automated click might appear from nowhere and cause an flurry of unlikely hover, focus events.
Only way it would be possible is with some specialist hardware and software on the mouse itself that sends evidence of the actual mechanical click. Via software this is not possible.
Although in my other answer I mention that you should ideally just respond to clicks as clicks, there is one possibility that could work, depending on how a "programmatic" click is generated.
I am assuming a Windows platform, due to the "vb.net" tag:
With the WinAPI you can send a message to any window in order to simulate, for example, a WM_LBUTTONDOWN event. In this message you would include the X and Y location of the mouse at the time the button was pressed - or where the receiving program expects it to be. When you handle the message you could use the GetCursorPos call to get the actual cursor position. Verify that the current position is close to that in the message, and handle it as a click, otherwise ignore it.
Bear in mind however that the nature of the message queue is such that it could take some time to handle the event, and the mouse can move a long way in a short space of time.
This solution would only work if the "click" is generated by a simple Send/PostMessage. If the application that is generating the click simulates the movement of the mouse also, then you should probably see the other answers :)
You can check the mouse event flags LLMHF_INJECTED and LLMHF_LOWER_IL_INJECTED or utilize input hooks to monitor mouse input data, where malformed data may indicated that input was injected.
https://learn.microsoft.com/en-us/windows/win32/api/winuser/ns-winuser-msllhookstruct
https://learn.microsoft.com/en-us/windows/win32/api/winuser/nf-winuser-setwindowshookexa
I'm not sure there is a way to determine with perfect accuracy whether a mouse click is automated or not. I mean, you could write your own USB driver that sits in between the native mouse driver and the OS and relays only "real" clicks to you. However, even that can be defeated by plugging in a USB device (like a smartphone) that's programmed to send USB packets to host computer.
What are you trying to accomplish that requires you to distinguish between real mouse clicks and fake ones?
Create a statistical learning solution by logging the past X mouse events in your program. When the user clicks on the control, determine the probability based on the last X actions that it's a real click.
Train your solution using real clicks and a large variety of automated scenarios.
This is obviously not a guaranteed-to-work solution and is more for fun than anything else.
It is much more harder than what you think because input macro programs produce legit mouse and keyboard input messages to your game. I don't think there is a way to check if the input message is actually triggered by a physical hardware input (like mouse or keyboard) unless OS provides you with accessibility of input-driver-level.
Since this is specifically for a game, you can see how other games handle this situation. Some of the common methods are,
Check frequency of mouse clicks. (human beings cannot click as fast as programs.)
At random points or when it doubts, use CAPTCHA to verify. (Read this: http://www.threadmeters.com/v-1Vvd/CAPTCHA_The_Obvious_AntiBot_Solution/ )
Use outside monitoring tools to inspect all processes running in a machine to find out programs known for cheating purpose. Steam does this. Check Valve Anti-Cheat System Blizzard's WoW also does the same thing with Warden.
My own advice would be "Use your gameplay system". Since every game has its own rules and gameplay styles, it wouldn't be too hard to detect whether a player is cheating or not. This approach won't be a general solution and it could be silly but if it works for your game, why not? :)
There are two ways you could work around this problem.
Make a new Button with a Text proving that the user isn't AFK and if the user doesn't press the button, kick him or her out. Generate the button in random locations.
If the user isn't moving for a while, (for example, 10 mins) kick him or her out.