How do I get the coordinates of a mouse click on a canvas element?


What's the simplest way to add a click event handler to a canvas element that will return the x and y coordinates of the click (relative to the canvas element)?

No legacy browser compatibility required, Safari, Opera and Firefox will do.

Edit 2018: This answer is pretty old and it uses checks for old browsers that are not necessary anymore, as the clientX and clientY properties work in all current browsers. You might want to check out Patriques Answer for a simpler, more recent solution.

Original Answer:
As described in an article i found back then but exists no longer:

var x;
var y;
if (e.pageX || e.pageY) { 
  x = e.pageX;
  y = e.pageY;
}
else { 
  x = e.clientX + document.body.scrollLeft + document.documentElement.scrollLeft; 
  y = e.clientY + document.body.scrollTop + document.documentElement.scrollTop; 
} 
x -= gCanvasElement.offsetLeft;
y -= gCanvasElement.offsetTop;

Worked perfectly fine for me.


If you like simplicity but still want cross-browser functionality I found this solution worked best for me. This is a simplification of @Aldekein´s solution but without jQuery.

function getCursorPosition(canvas, event) {
    const rect = canvas.getBoundingClientRect()
    const x = event.clientX - rect.left
    const y = event.clientY - rect.top
    console.log("x: " + x + " y: " + y)
}

const canvas = document.querySelector('canvas')
canvas.addEventListener('mousedown', function(e) {
    getCursorPosition(canvas, e)
})

Update (5/5/16): patriques' answer should be used instead, as it's both simpler and more reliable.


Since the canvas isn't always styled relative to the entire page, the canvas.offsetLeft/Top doesn't always return what you need. It will return the number of pixels it is offset relative to its offsetParent element, which can be something like a div element containing the canvas with a position: relative style applied. To account for this you need to loop through the chain of offsetParents, beginning with the canvas element itself. This code works perfectly for me, tested in Firefox and Safari but should work for all.

function relMouseCoords(event){
    var totalOffsetX = 0;
    var totalOffsetY = 0;
    var canvasX = 0;
    var canvasY = 0;
    var currentElement = this;

    do{
        totalOffsetX += currentElement.offsetLeft - currentElement.scrollLeft;
        totalOffsetY += currentElement.offsetTop - currentElement.scrollTop;
    }
    while(currentElement = currentElement.offsetParent)

    canvasX = event.pageX - totalOffsetX;
    canvasY = event.pageY - totalOffsetY;

    return {x:canvasX, y:canvasY}
}
HTMLCanvasElement.prototype.relMouseCoords = relMouseCoords;

The last line makes things convenient for getting the mouse coordinates relative to a canvas element. All that's needed to get the useful coordinates is

coords = canvas.relMouseCoords(event);
canvasX = coords.x;
canvasY = coords.y;

Modern browser's now handle this for you. Chrome, IE9, and Firefox support the offsetX/Y like this, passing in the event from the click handler.

function getRelativeCoords(event) {
    return { x: event.offsetX, y: event.offsetY };
}

Most modern browsers also support layerX/Y, however Chrome and IE use layerX/Y for the absolute offset of the click on the page including margin, padding, etc. In Firefox, layerX/Y and offsetX/Y are equivalent, but offset didn't previously exist. So, for compatibility with slightly older browsers, you can use:

function getRelativeCoords(event) {
    return { x: event.offsetX || event.layerX, y: event.offsetY || event.layerY };
}

According to fresh Quirksmode the clientX and clientY methods are supported in all major browsers. So, here it goes - the good, working code that works in a scrolling div on a page with scrollbars:

function getCursorPosition(canvas, event) {
var x, y;

canoffset = $(canvas).offset();
x = event.clientX + document.body.scrollLeft + document.documentElement.scrollLeft - Math.floor(canoffset.left);
y = event.clientY + document.body.scrollTop + document.documentElement.scrollTop - Math.floor(canoffset.top) + 1;

return [x,y];
}

This also requires jQuery for $(canvas).offset().


I made a full demostration that works in every browser with the full source code of the solution of this problem: Coordinates of a mouse click on Canvas in Javascript. To try the demo, copy the code and paste it into a text editor. Then save it as example.html and, finally, open the file with a browser.


Here is a small modification to Ryan Artecona's answer for canvases with a variable (%) width:

 HTMLCanvasElement.prototype.relMouseCoords = function (event) {
    var totalOffsetX = 0;
    var totalOffsetY = 0;
    var canvasX = 0;
    var canvasY = 0;
    var currentElement = this;

    do {
        totalOffsetX += currentElement.offsetLeft;
        totalOffsetY += currentElement.offsetTop;
    }
    while (currentElement = currentElement.offsetParent)

    canvasX = event.pageX - totalOffsetX;
    canvasY = event.pageY - totalOffsetY;

    // Fix for variable canvas width
    canvasX = Math.round( canvasX * (this.width / this.offsetWidth) );
    canvasY = Math.round( canvasY * (this.height / this.offsetHeight) );

    return {x:canvasX, y:canvasY}
}

Be wary while doing the coordinate conversion; there are multiple non-cross-browser values returned in a click event. Using clientX and clientY alone are not sufficient if the browser window is scrolled (verified in Firefox 3.5 and Chrome 3.0).

This quirks mode article provides a more correct function that can use either pageX or pageY or a combination of clientX with document.body.scrollLeft and clientY with document.body.scrollTop to calculate the click coordinate relative to the document origin.

UPDATE: Additionally, offsetLeft and offsetTop are relative to the padded size of the element, not the interior size. A canvas with the padding: style applied will not report the top-left of its content region as offsetLeft. There are various solutions to this problem; the simplest one may be to clear all border, padding, etc. styles on the canvas itself and instead apply them to a box containing the canvas.


I'm not sure what's the point of all these answers that loop through parent elements and do all kinds of weird stuff.

The HTMLElement.getBoundingClientRect method is designed to to handle actual screen position of any element. This includes scrolling, so stuff like scrollTop is not needed:

(from MDN) The amount of scrolling that has been done of the viewport area (or any other scrollable element) is taken into account when computing the bounding rectangle

Normal image

The very simplest approach was already posted here. This is correct as long as no wild CSS rules are involved.

Handling stretched canvas/image

When image pixel width isn't matched by it's CSS width, you'll need to apply some ratio on pixel values:

/* Returns pixel coordinates according to the pixel that's under the mouse cursor**/
HTMLCanvasElement.prototype.relativeCoords = function(event) {
  var x,y;
  //This is the current screen rectangle of canvas
  var rect = this.getBoundingClientRect();
  var top = rect.top;
  var bottom = rect.bottom;
  var left = rect.left;
  var right = rect.right;
  //Recalculate mouse offsets to relative offsets
  x = event.clientX - left;
  y = event.clientY - top;
  //Also recalculate offsets of canvas is stretched
  var width = right - left;
  //I use this to reduce number of calculations for images that have normal size 
  if(this.width!=width) {
    var height = bottom - top;
    //changes coordinates by ratio
    x = x*(this.width/width);
    y = y*(this.height/height);
  } 
  //Return as an array
  return [x,y];
}

As long as the canvas has no border, it works for stretched images (jsFiddle).

Handling CSS borders

If the canvas has thick border, the things get little complicated. You'll literally need to subtract the border from the bounding rectangle. This can be done using .getComputedStyle. This answer describes the process.

The function then grows up a little:

/* Returns pixel coordinates according to the pixel that's under the mouse cursor**/
HTMLCanvasElement.prototype.relativeCoords = function(event) {
  var x,y;
  //This is the current screen rectangle of canvas
  var rect = this.getBoundingClientRect();
  var top = rect.top;
  var bottom = rect.bottom;
  var left = rect.left;
  var right = rect.right;
  //Subtract border size
  // Get computed style
  var styling=getComputedStyle(this,null);
  // Turn the border widths in integers
  var topBorder=parseInt(styling.getPropertyValue('border-top-width'),10);
  var rightBorder=parseInt(styling.getPropertyValue('border-right-width'),10);
  var bottomBorder=parseInt(styling.getPropertyValue('border-bottom-width'),10);
  var leftBorder=parseInt(styling.getPropertyValue('border-left-width'),10);
  //Subtract border from rectangle
  left+=leftBorder;
  right-=rightBorder;
  top+=topBorder;
  bottom-=bottomBorder;
  //Proceed as usual
  ...
}

I can't think of anything that would confuse this final function. See yourself at JsFiddle.

Notes

If you don't like modifying the native prototypes, just change the function and call it with (canvas, event) (and replace any this with canvas).


Here is a very nice tutorial-

http://www.html5canvastutorials.com/advanced/html5-canvas-mouse-coordinates/

 <canvas id="myCanvas" width="578" height="200"></canvas>
<script>
  function writeMessage(canvas, message) {
    var context = canvas.getContext('2d');
    context.clearRect(0, 0, canvas.width, canvas.height);
    context.font = '18pt Calibri';
    context.fillStyle = 'black';
    context.fillText(message, 10, 25);
  }
  function getMousePos(canvas, evt) {
    var rect = canvas.getBoundingClientRect();
    return {
      x: evt.clientX - rect.left,
      y: evt.clientY - rect.top
    };
  }
  var canvas = document.getElementById('myCanvas');
  var context = canvas.getContext('2d');

  canvas.addEventListener('mousemove', function(evt) {
    var mousePos = getMousePos(canvas, evt);
    var message = 'Mouse position: ' + mousePos.x + ',' + mousePos.y;
    writeMessage(canvas, message);
  }, false);

hope this helps!


Using jQuery in 2016, to get click coordinates relative to the canvas, I do:

$(canvas).click(function(jqEvent) {
    var coords = {
        x: jqEvent.pageX - $(canvas).offset().left,
        y: jqEvent.pageY - $(canvas).offset().top
    };
});

This works since both canvas offset() and jqEvent.pageX/Y are relative to the document regardless of scroll position.

Note that if your canvas is scaled then these coordinates are not the same as canvas logical coordinates. To get those, you would also do:

var logicalCoords = {
    x: coords.x * (canvas.width / $(canvas).width()),
    y: coords.y * (canvas.height / $(canvas).height())
}

So this is both simple but a slightly more complicated topic than it seems.

First off there are usually to conflated questions here

  1. How to get element relative mouse coordinates

  2. How to get canvas pixel mouse coordinates for the 2D Canvas API or WebGL

so, answers

How to get element relative mouse coordinates

Whether or not the element is a canvas getting element relative mouse coordinates is the same for all elements.

There are 2 simple answers to the question "How to get canvas relative mouse coordinates"

Simple answer #1 use offsetX and offsetY

canvas.addEventListner('mousemove', (e) => {
  const x = e.offsetX;
  const y = e.offsetY;
});

This answer works in Chrome, Firefox, and Safari. Unlike all the other event values offsetX and offsetY take CSS transforms into account.

The biggest problem with offsetX and offsetY is as of 2019/05 they don't exist on touch events and so can't be used with iOS Safari. They do exist on Pointer Events which exist in Chrome and Firefox but not Safari although apparently Safari is working on it.

Another issue is the events must be on the canvas itself. If you put them on some other element or the window you can not later choose the canvas to be your point of reference.

Simple answer #2 use clientX, clientY and canvas.getBoundingClientRect

If you don't care about CSS transforms the next simplest answer is to call canvas. getBoundingClientRect() and subtract the left from clientX and top from clientY as in

canvas.addEventListener('mousemove', (e) => {
  const rect = canvas.getBoundingClientRect();
  const x = e.clientX - rect.left;
  const y = e.clientY - rect.top;
});

This will work as long as there are no CSS transforms. It also works with touch events and so will work with Safari iOS

canvas.addEventListener('touchmove', (e) => {
  const rect = canvas. getBoundingClientRect();
  const x = e.touches[0].clientX - rect.left;
  const y = e.touches[0].clientY - rect.top;
});

How to get canvas pixel mouse coordinates for the 2D Canvas API

For this we need to take the values we got above and convert from the size the canvas is displayed to the number of pixels in the canvas itself

with canvas.getBoundingClientRect and clientX and clientY

canvas.addEventListener('mousemove', (e) => {
  const rect = canvas.getBoundingClientRect();
  const elementRelativeX = e.clientX - rect.left;
  const elementRelativeY = e.clientY - rect.top;
  const canvasRelativeX = elementRelativeX * canvas.width / rect.width;
  const canvasRelativeY = elementRelativeY * canvas.height / rect.height;
});

or with offsetX and offsetY

canvas.addEventListener('mousemove', (e) => {
  const elementRelativeX = e.offsetX;
  const elementRelativeX = e.offsetY;
  const canvasRelativeX = elementRelativeX * canvas.width / canvas.clientWidth;
  const canvasRelativeY = elementRelativeX * canvas.height / canvas.clientHeight;
});

Note: In all cases do not add padding or borders to the canvas. Doing so will massively complicate the code. Instead of you want a border or padding surround the canvas in some other element and add the padding and or border to the outer element.

Working example using event.offsetX, event.offsetY

[...document.querySelectorAll('canvas')].forEach((canvas) => {
  const ctx = canvas.getContext('2d');
  ctx.canvas.width  = ctx.canvas.clientWidth;
  ctx.canvas.height = ctx.canvas.clientHeight;
  let count = 0;

  function draw(e, radius = 1) {
    const pos = {
      x: e.offsetX * canvas.width  / canvas.clientWidth,
      y: e.offsetY * canvas.height / canvas.clientHeight,
    };
    document.querySelector('#debug').textContent = count;
    ctx.beginPath();
    ctx.arc(pos.x, pos.y, radius, 0, Math.PI * 2);
    ctx.fillStyle = hsl((count++ % 100) / 100, 1, 0.5);
    ctx.fill();
  }

  function preventDefault(e) {
    e.preventDefault();
  }

  if (window.PointerEvent) {
    canvas.addEventListener('pointermove', (e) => {
      draw(e, Math.max(Math.max(e.width, e.height) / 2, 1));
    });
    canvas.addEventListener('touchstart', preventDefault, {passive: false});
    canvas.addEventListener('touchmove', preventDefault, {passive: false});
  } else {
    canvas.addEventListener('mousemove', draw);
    canvas.addEventListener('mousedown', preventDefault);
  }
});

function hsl(h, s, l) {
  return `hsl(${h * 360 | 0},${s * 100 | 0}%,${l * 100 | 0}%)`;
}
.scene {
  width: 200px;
  height: 200px;
  perspective: 600px;
}

.cube {
  width: 100%;
  height: 100%;
  position: relative;
  transform-style: preserve-3d;
  animation-duration: 16s;
  animation-name: rotate;
  animation-iteration-count: infinite;
  animation-timing-function: linear;
}

@keyframes rotate {
  from { transform: translateZ(-100px) rotateX(  0deg) rotateY(  0deg); }
  to   { transform: translateZ(-100px) rotateX(360deg) rotateY(720deg); }
}

.cube__face {
  position: absolute;
  width: 200px;
  height: 200px;
  display: block;
}

.cube__face--front  { background: rgba(255, 0, 0, 0.2); transform: rotateY(  0deg) translateZ(100px); }
.cube__face--right  { background: rgba(0, 255, 0, 0.2); transform: rotateY( 90deg) translateZ(100px); }
.cube__face--back   { background: rgba(0, 0, 255, 0.2); transform: rotateY(180deg) translateZ(100px); }
.cube__face--left   { background: rgba(255, 255, 0, 0.2); transform: rotateY(-90deg) translateZ(100px); }
.cube__face--top    { background: rgba(0, 255, 255, 0.2); transform: rotateX( 90deg) translateZ(100px); }
.cube__face--bottom { background: rgba(255, 0, 255, 0.2); transform: rotateX(-90deg) translateZ(100px); }
<div class="scene">
  <div class="cube">
    <canvas class="cube__face cube__face--front"></canvas>
    <canvas class="cube__face cube__face--back"></canvas>
    <canvas class="cube__face cube__face--right"></canvas>
    <canvas class="cube__face cube__face--left"></canvas>
    <canvas class="cube__face cube__face--top"></canvas>
    <canvas class="cube__face cube__face--bottom"></canvas>
  </div>
</div>
<pre id="debug"></pre>

Working example using canvas.getBoundingClientRect and event.clientX and event.clientY

const canvas = document.querySelector('canvas');
const ctx = canvas.getContext('2d');
ctx.canvas.width  = ctx.canvas.clientWidth;
ctx.canvas.height = ctx.canvas.clientHeight;
let count = 0;

function draw(e, radius = 1) {
  const rect = canvas.getBoundingClientRect();
  const pos = {
    x: (e.clientX - rect.left) * canvas.width  / canvas.clientWidth,
    y: (e.clientY - rect.top) * canvas.height / canvas.clientHeight,
  };
  ctx.beginPath();
  ctx.arc(pos.x, pos.y, radius, 0, Math.PI * 2);
  ctx.fillStyle = hsl((count++ % 100) / 100, 1, 0.5);
  ctx.fill();
}

function preventDefault(e) {
  e.preventDefault();
}

if (window.PointerEvent) {
  canvas.addEventListener('pointermove', (e) => {
    draw(e, Math.max(Math.max(e.width, e.height) / 2, 1));
  });
  canvas.addEventListener('touchstart', preventDefault, {passive: false});
  canvas.addEventListener('touchmove', preventDefault, {passive: false});
} else {
  canvas.addEventListener('mousemove', draw);
  canvas.addEventListener('mousedown', preventDefault);
}

function hsl(h, s, l) {
  return `hsl(${h * 360 | 0},${s * 100 | 0}%,${l * 100 | 0}%)`;
}
canvas { background: #FED; }
<canvas width="400" height="100" style="width: 300px; height: 200px"></canvas>
<div>canvas deliberately has differnt CSS size vs drawingbuffer size</div>


I recommend this link- http://miloq.blogspot.in/2011/05/coordinates-mouse-click-canvas.html

<style type="text/css">

  #canvas{background-color: #000;}

</style>

<script type="text/javascript">

  document.addEventListener("DOMContentLoaded", init, false);

  function init()
  {
    var canvas = document.getElementById("canvas");
    canvas.addEventListener("mousedown", getPosition, false);
  }

  function getPosition(event)
  {
    var x = new Number();
    var y = new Number();
    var canvas = document.getElementById("canvas");

    if (event.x != undefined && event.y != undefined)
    {
      x = event.x;
      y = event.y;
    }
    else // Firefox method to get the position
    {
      x = event.clientX + document.body.scrollLeft +
          document.documentElement.scrollLeft;
      y = event.clientY + document.body.scrollTop +
          document.documentElement.scrollTop;
    }

    x -= canvas.offsetLeft;
    y -= canvas.offsetTop;

    alert("x: " + x + "  y: " + y);
  }

</script>

In Prototype, use cumulativeOffset() to do the recursive summation as mentioned by Ryan Artecona above.

http://www.prototypejs.org/api/element/cumulativeoffset


You could just do:

var canvas = yourCanvasElement;
var mouseX = (event.clientX - (canvas.offsetLeft - canvas.scrollLeft)) - 2;
var mouseY = (event.clientY - (canvas.offsetTop - canvas.scrollTop)) - 2;

This will give you the exact position of the mouse pointer.


See demo at http://jsbin.com/ApuJOSA/1/edit?html,output .

  function mousePositionOnCanvas(e) {
      var el=e.target, c=el;
      var scaleX = c.width/c.offsetWidth || 1;
      var scaleY = c.height/c.offsetHeight || 1;

      if (!isNaN(e.offsetX)) 
          return { x:e.offsetX*scaleX, y:e.offsetY*scaleY };

      var x=e.pageX, y=e.pageY;
      do {
        x -= el.offsetLeft;
        y -= el.offsetTop;
        el = el.offsetParent;
      } while (el);
      return { x: x*scaleX, y: y*scaleY };
  }

Here is some modifications of the above Ryan Artecona's solution.

function myGetPxStyle(e,p)
{
    var r=window.getComputedStyle?window.getComputedStyle(e,null)[p]:"";
    return parseFloat(r);
}

function myGetClick=function(ev)
{
    // {x:ev.layerX,y:ev.layerY} doesn't work when zooming with mac chrome 27
    // {x:ev.clientX,y:ev.clientY} not supported by mac firefox 21
    // document.body.scrollLeft and document.body.scrollTop seem required when scrolling on iPad
    // html is not an offsetParent of body but can have non null offsetX or offsetY (case of wordpress 3.5.1 admin pages for instance)
    // html.offsetX and html.offsetY don't work with mac firefox 21

    var offsetX=0,offsetY=0,e=this,x,y;
    var htmls=document.getElementsByTagName("html"),html=(htmls?htmls[0]:0);

    do
    {
        offsetX+=e.offsetLeft-e.scrollLeft;
        offsetY+=e.offsetTop-e.scrollTop;
    } while (e=e.offsetParent);

    if (html)
    {
        offsetX+=myGetPxStyle(html,"marginLeft");
        offsetY+=myGetPxStyle(html,"marginTop");
    }

    x=ev.pageX-offsetX-document.body.scrollLeft;
    y=ev.pageY-offsetY-document.body.scrollTop;
    return {x:x,y:y};
}

First, as others have said, you need a function to get the position of the canvas element. Here's a method that's a little more elegant than some of the others on this page (IMHO). You can pass it any element and get its position in the document:

function findPos(obj) {
    var curleft = 0, curtop = 0;
    if (obj.offsetParent) {
        do {
            curleft += obj.offsetLeft;
            curtop += obj.offsetTop;
        } while (obj = obj.offsetParent);
        return { x: curleft, y: curtop };
    }
    return undefined;
}

Now calculate the current position of the cursor relative to that:

$('#canvas').mousemove(function(e) {
    var pos = findPos(this);
    var x = e.pageX - pos.x;
    var y = e.pageY - pos.y;
    var coordinateDisplay = "x=" + x + ", y=" + y;
    writeCoordinateDisplay(coordinateDisplay);
});

Notice that I've separated the generic findPos function from the event handling code. (As it should be. We should try to keep our functions to one task each.)

The values of offsetLeft and offsetTop are relative to offsetParent, which could be some wrapper div node (or anything else, for that matter). When there is no element wrapping the canvas they're relative to the body, so there is no offset to subtract. This is why we need to determine the position of the canvas before we can do anything else.

Similary, e.pageX and e.pageY give the position of the cursor relative to the document. That's why we subtract the canvas's offset from those values to arrive at the true position.

An alternative for positioned elements is to directly use the values of e.layerX and e.layerY. This is less reliable than the method above for two reasons:

  1. These values are also relative to the entire document when the event does not take place inside a positioned element
  2. They are not part of any standard

ThreeJS r77

var x = event.offsetX == undefined ? event.layerX : event.offsetX;
var y = event.offsetY == undefined ? event.layerY : event.offsetY;

mouse2D.x = ( x / renderer.domElement.width ) * 2 - 1;
mouse2D.y = - ( y / renderer.domElement.height ) * 2 + 1;

After trying many solutions. This worked for me. Might help someone else hence posting. Got it from here


Here is a simplified solution (this doesn't work with borders/scrolling):

function click(event) {
    const bound = event.target.getBoundingClientRect();

    const xMult = bound.width / can.width;
    const yMult = bound.height / can.height;

    return {
        x: Math.floor(event.offsetX / xMult),
        y: Math.floor(event.offsetY / yMult),
    };
}

I was creating an application having a canvas over a pdf, that involved a lot of resizes of canvas like Zooming the pdf-in and out, and in turn on every zoom-in/out of PDF I had to resize the canvas to adapt the size of the pdf, I went through lot of answers in stackOverflow, and didn't found a perfect solution that will eventually solve the problem.

I was using rxjs and angular 6, and didn't found any answer specific to the newest version.

Here is the entire code snippet that would be helpful, to anyone leveraging rxjs to draw on top of canvas.

  private captureEvents(canvasEl: HTMLCanvasElement) {

    this.drawingSubscription = fromEvent(canvasEl, 'mousedown')
      .pipe(
        switchMap((e: any) => {

          return fromEvent(canvasEl, 'mousemove')
            .pipe(
              takeUntil(fromEvent(canvasEl, 'mouseup').do((event: WheelEvent) => {
                const prevPos = {
                  x: null,
                  y: null
                };
              })),

              takeUntil(fromEvent(canvasEl, 'mouseleave')),
              pairwise()
            )
        })
      )
      .subscribe((res: [MouseEvent, MouseEvent]) => {
        const rect = this.cx.canvas.getBoundingClientRect();
        const prevPos = {
          x: Math.floor( ( res[0].clientX - rect.left ) / ( rect.right - rect.left ) * this.cx.canvas.width ),
          y:  Math.floor( ( res[0].clientY - rect.top ) / ( rect.bottom - rect.top ) * this.cx.canvas.height )
        };
        const currentPos = {
          x: Math.floor( ( res[1].clientX - rect.left ) / ( rect.right - rect.left ) * this.cx.canvas.width ),
          y: Math.floor( ( res[1].clientY - rect.top ) / ( rect.bottom - rect.top ) * this.cx.canvas.height )
        };

        this.coordinatesArray[this.file.current_slide - 1].push(prevPos);
        this.drawOnCanvas(prevPos, currentPos);
      });
  }

And here is the snippet that fixes, mouse coordinates relative to size of the canvas, irrespective of how you zoom-in/out the canvas.

const prevPos = {
  x: Math.floor( ( res[0].clientX - rect.left ) / ( rect.right - rect.left ) * this.cx.canvas.width ),
  y:  Math.floor( ( res[0].clientY - rect.top ) / ( rect.bottom - rect.top ) * this.cx.canvas.height )
};
const currentPos = {
  x: Math.floor( ( res[1].clientX - rect.left ) / ( rect.right - rect.left ) * this.cx.canvas.width ),
  y: Math.floor( ( res[1].clientY - rect.top ) / ( rect.bottom - rect.top ) * this.cx.canvas.height )
};

Hey, this is in dojo, just cause it's what I had the code in already for a project.

It should be fairly Obvious how to convert it back to non dojo vanilla JavaScript.

  function onMouseClick(e) {
      var x = e.clientX;
      var y = e.clientY;
  }
  var canvas = dojo.byId(canvasId);
  dojo.connect(canvas,"click",onMouseClick);

Hope that helps.