Detect when MousePosition in a RectTransform

Hello Unity Community,

in my 2d game I am trying to find out when is Input.mousePosition on a UI Image. I want to do that by checking out if Input.mousePosition coordinates are inside the borders given by RectTransform.rect.xMin, .xMax, .yMin, .yMax:

if (Input.mousePosition.x < (imgRectTransform.position.x + imgRectTransform.rect.xMax)
&& Input.mousePosition.x > (imgRectTransform.position.x + imgRectTransform.rect.xMin)
&& Input.mousePosition.y <(imgRectTransform.position.y + imgRectTransform.rect.yMax)
&& Input.mousePosition.y > (imgRectTransform.position.y + imgRectTransform.rect.yMin))
return true;

This boolean expression is true when I click over the image, but it is also true when i click "quite a bit" outside of the image. "Quite a bit" is, as testing shows me, depending on the screen size - from a couple of pixels for larger screen to double size of the Image for smaller screens.. And I don't know why exactly?..

What I figured just out - it has to do with with the canvasScaler - which is set to "Scale With Screen Size" with Full HD as reference resolution and "Match Width Or Height" = 1. Canvas.scaleFactor gives me 1.. So what am I missing in the calculation?

Any opinions and suggestions are very welcome!

 

No need to check that yourself, check out https://docs.unity3d.com/ScriptReference/EventSystems.IPointerEnterHandler.html

 

First of all - thank you for the answer, @Johannski

For this to work (which really works), the event system and Graphic Raycaster must be enabled. For performance reasons I don't want the Raycaster to be enabled. In other words, checking if a click is on button only when a touch / mouse click is made with a simple boolean comparison of floats is more performant then letting a raycaster spawn a ray every frame to all object that are raycast targets (in my UI I have more buttons than the mentioned one). Also I do cache all the values to be checked in temp variables -> no new objects created internally -> GC called rarely.

 

Have a look at: https://docs.unity3d.com/ScriptRefe...ility.ScreenPointToLocalPointInRectangle.html

Might help you.

 

I don't think that is what Tetsubo wants to use either if performance is important. Have a look at the decompiled version: https://github.com/MattRix/UnityDec...ityEngine/UnityEngine/RectTransformUtility.cs

It creates rays (garbage) and tests those rays, thus it is not that different if not worse than using a graphics raycaster.

Maybe try this instead:

Code (CSharp):

1. var mousePosition = Input.mousePosition;
2.             var normalizedMousePosition = new Vector2(mousePosition.x / Screen.width, mousePosition.y / Screen.height);
3.             if (normalizedMousePosition.x > rectTransform.anchorMin.x &&
4.                 normalizedMousePosition.x < rectTransform.anchorMax.x &&
5.                 normalizedMousePosition.y > rectTransform.anchorMin.y &&
6.                 normalizedMousePosition.y < rectTransform.anchorMax.y)
7.             {
8.                 // Hit the rect transform
9.             }

 

@jeffweber , thanks mate! That works pretty well and is the shortest way to that kind of checks.. as @Johannski pointer out though, this is not that performant as it creates quite a bit of garbage..

I will test tomorrow the suggested solution by @Johannski , it looks really very promising! Thanks a lot!

 

This works for me:

Code (CSharp):

1. Vector2 localMousePosition = imgRectTransform.InverseTransformPoint(Input.mousePosition);
2.         if (imgRectTransform.rect.Contains(localMousePosition))
3.         {
4.             return true;
5.         }

 

THANKYOU!! This works with Input.GetTouch(0).position as well, exactly what I needed.
Legend mate, thanks.

 

Legendary tip, even 3 years into the future. Thank mate

 

You can also use the following, which is helpful in case you change the camera used:
Camera.main.ScreenToWorldPoint(new Vector3(touch.position.x, touch.position.y, Camera.main.farClipPlane));

 

For Screen Space - Overlay:
RectTransformUtility.RectangleContainsScreenPoint(wordRect, Input.mousePosition);

For Screen Space - Camera:
RectTransformUtility.RectangleContainsScreenPoint(wordRect, Input.mousePosition, Camera.main);
or the camera that you used.

 

It didn't quite work for me but my situation is maybe a bit different. I need to check if the mouse is over a button (thus having a RectTransform) but due to all sorts of changes and filtering events on the canvas the button itself is not clickable with OnClick since it's Image component is turned off (but still on the object). Because of that, I couldn't use all the IsPointerEnter and his friends and had to go more basic.

Also, I don't use Overlay or World camera so this is probably why the method above doesn't work. This is how to do it with Screen Space Camera canvas:

Code (CSharp):

1.            
2. Vector2 localPoint;
3. RectTransformUtility.ScreenPointToLocalPointInRectangle(_rectTransfiorm, Input.mousePosition, Camera.main, out localPoint);
4.             return _rectTransfiorm.rect.Contains(localPoint);

 

This code works beautifully but is there a way to give some "padding" to the recttransfrom in this code? Meaning something like: if mouse is within rectangle bounds + 100px padding

 

FYI rays are not "garbage" that need to be GCed. As they are struct they are copied on the stack instead of being allocated in the heap. Like Vector3, Vector3, Rectangle, etc..

 

So clean. So smooth. Thank you