I have an image which I want to crop.
Thus, I created canvas as overlay with a darker background and then on mouse move I draw an rect on that canvas and in that rect I clear the dark background. That looks as follows:
Thus the area without background will be cropped.
My code is as follows:
var canvas = document.getElementById('canvas');
var img = document.getElementById('photo');
var ctx = canvas.getContext('2d');
var rect = {};
var drag = false;
var update = true; // when true updates canvas
var original_source = img.src;
img.src = original_source;
var deg = 0;
var image_rotate_angle = 90;
document.getElementById('right').addEventListener("click", rotateRight, false);
document.getElementById('left').addEventListener("click", rotateLeft, false);
function rotateRight(){
deg = deg + image_rotate_angle;
img.style.transform = "rotate(" + deg + "deg)";
canvas.style.transform = "rotate(" + deg + "deg)";
}
function rotateLeft(){
deg = deg - image_rotate_angle;
img.style.transform = "rotate(" + deg + "deg)";
canvas.style.transform = "rotate(" + deg + "deg)";
}
function init() {
img.addEventListener('load', function(){
canvas.width = img.width;
canvas.height = img.height;
canvas.addEventListener('mousedown', mouseDown, false);
canvas.addEventListener('mouseup', mouseUp, false);
canvas.addEventListener('mousemove', mouseMove, false);
});
// start the rendering loop
requestAnimationFrame(updateCanvas);
}
// main render loop only updates if update is true
function updateCanvas(){
if(update){
drawCanvas();
update = false;
}
requestAnimationFrame(updateCanvas);
}
// draws a rectangle with rotation
function drawRect(){
ctx.clearRect(rect.startX, rect.startY, rect.w, rect.h);
}
// clears canvas sets filters and draws rectangles
function drawCanvas(){
ctx.clearRect(0, 0, canvas.width, canvas.height);
ctx.fillStyle = 'rgba(32, 32, 32, 0.7)';
ctx.fillRect(0, 0, canvas.width, canvas.height);
drawRect()
}
// create new rect add to array
function mouseDown(e) {
rect = {
startX : e.offsetX,
startY : e.offsetY,
w : 1,
h : 1
};
drag = true;
}
function mouseUp() { drag = false; update = true; }
function mouseMove(e) {
if (drag) {
rect.w = (e.pageX - this.offsetLeft) - rect.startX;
rect.h = (e.pageY - this.offsetTop) - rect.startY;
update = true;
}
}
init();
canvas{
position: absolute;
left: 0;
right: 0;
top: 0;
bottom: 0;
display:inline-block;
background:rgba(0,0,0,0.3);
}
<div style="margin-bottom: 15px;">
<button id="left">Rotate Reft</button>
<button id="right">Rotate Right</button>
</div>
<div style="position: relative; overflow: hidden;display:inline-block;">
<img id="photo" src="http://www.html5canvastutorials.com/demos/assets/darth-vader.jpg"/>
<canvas id="canvas"></canvas>
</div>
I have also two buttons which I use to rotate the image. And there I have the problem. The solution is maybe obvious but I do not see it.
If I draw a the rectangle and then rotate the image, I rotate the image and the canvas also, and then rectangle stays at the same place after rotating. That is what I want and that works fine.
The problem is if I first rotate the image and then try to draw a rectangle. Then the rectangle isn't drawn as it should be, it goes in the opposite direction.
Here is the fiddle.
Any idea how to solve it?
You could probably just get the maths to find where the new x and y will lay, based on your image's size, but I guess it will be more manageable to actually draw everything on your canvas, and rotate its transformation matrix.
But this implies some changes in your code:
You would have to draw the image with compositing (or with 'even-odd' fill-rule, like I already shown you).
You would also have to make your drawCanvas function a bit more complicated in order to rotate your hole only when needed (when the user clicked the "rotate" buttons).
var canvas = document.getElementById('canvas');
var img = new Image();
var ctx = canvas.getContext('2d');
var rect = {};
var drag = false;
var update = true; // when true updates canvas
var angle = 0;
document.getElementById('right').addEventListener("click", rotateRight, false);
document.getElementById('left').addEventListener("click", rotateLeft, false);
function rotateRight() {
angle += Math.PI / 2;
drawCanvas(true); // let know we're from rotate function
}
function rotateLeft() {
angle -= Math.PI / 2;
drawCanvas(true);
}
function init() {
img.addEventListener('load', function() {
canvas.width = img.width;
canvas.height = img.height;
canvas.addEventListener('mousedown', mouseDown, false);
canvas.addEventListener('mouseup', mouseUp, false);
canvas.addEventListener('mousemove', mouseMove, false);
});
// start the rendering loop
requestAnimationFrame(updateCanvas);
}
// main render loop only updates if update is true
function updateCanvas() {
if (update) {
drawCanvas(false); // don't rotate the rectangle here
update = false;
}
requestAnimationFrame(updateCanvas);
}
// draws a rectangle with rotation
function drawRect() {
ctx.clearRect(rect.startX, rect.startY, rect.w, rect.h);
}
// clears canvas sets filters and draws rectangles
function drawCanvas(fromRotate) {
// reset the transformation matrix to its defaults
ctx.setTransform(1, 0, 0, 1, 0, 0);
ctx.clearRect(0, 0, canvas.width, canvas.height);
ctx.fillStyle = 'rgba(32, 32, 32, 0.7)';
ctx.fillRect(0, 0, canvas.width, canvas.height);
// for the rectangle, rotate only if we are using previously set rect
if (fromRotate) {
// Move to center
ctx.setTransform(1, 0, 0, 1, canvas.width / 2, canvas.height / 2);
// rotate by the current angle
ctx.rotate(angle);
// move back to what is now top right corner
ctx.translate(-canvas.width / 2, -canvas.height / 2);
}
drawRect();
// draw the image behind the overlay
ctx.globalCompositeOperation = 'destination-over';
// rotate the whole context every time for the image
ctx.setTransform(1, 0, 0, 1, canvas.width / 2, canvas.height / 2);
// rotate by the current angle
ctx.rotate(angle);
// draw the image, from center
ctx.drawImage(img, -img.width / 2, -img.height / 2);
}
// create new rect add to array
function mouseDown(e) {
rect = {
startX: e.offsetX,
startY: e.offsetY,
w: 1,
h: 1
};
drag = true;
}
function mouseUp() {
drag = false;
update = true;
}
function mouseMove(e) {
if (drag) {
rect.w = (e.offsetX - this.offsetLeft) - rect.startX;
rect.h = (e.offsetY - this.offsetTop) - rect.startY;
update = true;
}
}
init();
img.src = 'https://www.html5canvastutorials.com/demos/assets/darth-vader.jpg';
<div style="margin-bottom: 15px;">
<button id="left">Rotate Reft</button>
<button id="right">Rotate Right</button>
</div>
<div style="position: relative; overflow: hidden;display:inline-block;">
<canvas id="canvas"></canvas>
</div>
Related
I have a blue circle which is rotating around the red circle and moves on canvas continuously in one direction as long as the button is pressed.
Now I want to draw with the red circle while it is moving when the button is pressed (trace of its path).
Problems:
I have tried to make changes to clearRect() but I didn't succeed. the blue circle starts to draw on the canvas while moving which I don't need.
If its not possible to do with clearRect() function, Is it possible to do this by stacking canvas layers. Please help with example
const canvas = document.getElementById('canvas1');
const ctx = canvas.getContext('2d');
canvas.width = window.innerWidth;
canvas.height = window.innerHeight;
let positionX = 100;
let positionY = 100;
let X = 50;
let Y = 50;
let angle = 0;
let mouseButtonDown = false;
document.addEventListener('mousedown', () => mouseButtonDown = true);
document.addEventListener('mouseup', () => mouseButtonDown = false);
function circle(){
ctx.fillStyle = 'red';
ctx.beginPath();
ctx.arc(X, Y, 20, 0, Math.PI*2);
ctx.closePath();
ctx.fill();
}
function direction(){
ctx.fillStyle = 'blue';
ctx.beginPath();
ctx.arc(positionX + X, positionY + Y, 10, 0, Math.PI*2);
ctx.closePath();
positionX = 35 * Math.sin(angle);
positionY = 35 * Math.cos(angle);
ctx.fill();
}
function animate(){
if (mouseButtonDown) {
X += positionX / 10;
Y += positionY / 10;
} else {
angle += 0.1;
}
ctx.clearRect(X-positionX,Y-positionY, 20, 20);
circle();
direction();
requestAnimationFrame(animate);
}
animate();
#canvas1{
position: absolute;
top:0;
left: 0;
width: 100%;
height: 100%;
}
<!DOCTYPE html>
<html lang="en">
<head>
<link rel="stylesheet" href="style.css">
</head>
<body>
<canvas id="canvas1"></canvas>
<script src="script.js"></script>
</body>
</html>
Don`t stack canvas on the page
Each canvas you add to the page increases the amount of work the GPU and page compositor needs to do to render the page.
Use a second canvas that is not on the page and do the compositing by rendering the canvas to the onpage canvas using ctx.drawImage(secondCanvas, 0, 0).
This reduces the workload for the compositor, and in many cases avoid the need to do an addition image render (composite) for the second canvas I.E. onpage can require 3 drawImages (one for each canvas and once for the result) rather than 2 (once in your code and once as the result) if you use only one onpage canvas.
Using second canvas
Create a second canvas to store the drawn red lines.
You can create a copy of a canvas using
function copyCanvas(canvas, copyContent = false) {
const can = Object.assign(document.createElement("canvas"), {
width: canvas.width, height: canvas.height
});
can.ctx = can.getContext("2d");
copyContent && can.ctx.drawImage(canvas, 0, 0);
return can;
}
When you create render functions like circle, and direction pass as an argument the 2D context eg circle(ctx) so that it is easy to direct the rendering to any canvas.
function circle(ctx){
ctx.fillStyle = 'red';
ctx.beginPath();
ctx.arc(X, Y, redSize, 0, Math.PI*2);
ctx.fill();
}
// the background canvas
const bgCan = copyCanvas(canvas);
circle(bgCan.ctx); // will draw to the background canvas
Updating animation
When animating is is easiest to clear the whole canvas rather than mess about clearing only rendered pixels. Clearing rendered pixels gets complicated very quickly and will end up being many times slower than clearing the whole canvas.
After you clear the canvas draw the background canvas to the main canvas
ctx.clearRect(0, 0, ctx.canvas.width, ctx.canvas.height);
ctx.drawImage(bgCan, 0, 0);
When the mouse button is down draw the circle to the background canvas and while it is up draw to the main canvas.
Example
Adds a function to copy a canvas. copyCanvas
Clears the main canvas, and draws the background canvas onto the main canvas.
Render functions circle and direction have argument ctx to direct rendering to any context.
When mouse is down circle is drawn to background canvas bgCan else to the main canvas.
requestAnimationFrame(animate);
const ctx = canvas1.getContext('2d');
canvas1.width = innerWidth;
canvas1.height = innerHeight;
const bgCan = copyCanvas(canvas1);
const redSize = 10, blueSize = 5; // circle sizes on pixels
const drawSpeed = 2; // when button down draw speed in pixels per frame
var X = 50, Y = 50;
var angle = 0;
var mouseButtonDown = false;
document.addEventListener('mousedown', () => mouseButtonDown = true);
document.addEventListener('mouseup', () => mouseButtonDown = false);
function copyCanvas(canvas) {
const can = Object.assign(document.createElement("canvas"), {
width: canvas.width, height: canvas.height
});
can.ctx = can.getContext("2d");
return can;
}
function circle(ctx){
ctx.fillStyle = 'red';
ctx.beginPath();
ctx.arc(X, Y, redSize, 0, Math.PI*2);
ctx.fill();
}
function direction(ctx){
const d = blueSize + redSize + 5;
ctx.fillStyle = 'blue';
ctx.beginPath();
ctx.arc(d * Math.sin(angle) + X, d * Math.cos(angle) + Y, blueSize, 0, Math.PI*2);
ctx.fill();
}
function animate(){
ctx.clearRect(0, 0, ctx.canvas.width, ctx.canvas.height);
ctx.drawImage(bgCan, 0, 0);
if (mouseButtonDown) {
circle(bgCan.ctx);
X += Math.sin(angle) * drawSpeed;
Y += Math.cos(angle) * drawSpeed;
} else {
angle += 0.1;
circle(ctx);
}
direction(ctx);
requestAnimationFrame(animate);
}
#canvas1{
position: absolute;
top:0;
left: 0;
width: 100%;
height: 100%;
}
<canvas id="canvas1"></canvas>
BTW ctx.closePath() is like ctx.lineTo it is not the opposite to ctx.beginPath. A full arc or if you are just filling a shape you don't need to use ctx.closePath
BTW window is the default this, you don't need to include it, you dont use it to get at window.documentso why use it forwindow.innerWidth(same asinnerWidth` )
You could alter your code to keep track of the path of the red circle, with an array property, like this:
const canvas = document.getElementById('canvas1');
const ctx = canvas.getContext('2d');
canvas.width = window.innerWidth;
canvas.height = window.innerHeight;
let mouseButtonDown = false;
document.addEventListener('mousedown', () => mouseButtonDown = true);
document.addEventListener('mouseup', () => mouseButtonDown = false);
function drawCircle({x, y, radius, color}) {
ctx.fillStyle = color;
ctx.beginPath();
ctx.arc(x, y, radius, 0, Math.PI*2);
ctx.fill();
}
const red = { x: 50, y: 50, radius: 20, color: "red", path: [] };
const blue = { x: 100, y: 100, radius: 10, color: "blue", angle: 0 };
function animate(){
if (mouseButtonDown) {
red.path.push({x: red.x, y: red.y}); // store the old value
red.x += (blue.x - red.x) / 10;
red.y += (blue.y - red.y) / 10;
} else {
blue.angle += 0.1;
}
blue.x = red.x + 35 * Math.sin(blue.angle);
blue.y = red.y + 35 * Math.cos(blue.angle);
ctx.clearRect(0, 0, canvas.width, canvas.height); // clear the whole canvas
for (const {x, y} of red.path) { // draw circle at all the previous positions
drawCircle({...red, x, y});
}
drawCircle(red);
drawCircle(blue);
requestAnimationFrame(animate);
}
animate();
Using 2 canvases also works and may perform better especially when the path of the red circle has gotten long, because the background canvas doesn't need to be cleared and redrawn. Add a 2nd canvas in your html page with the same positioning, and give them ids 'background' and 'foreground'. You can then adjust the code to draw the blue circle to the foreground and red circles to the background (or vice versa).
// Create 2 canvases, set them to full size and get the contexts
const backgroundCanvas = document.getElementById('background');
const foregroundCanvas = document.getElementById('foreground');
const background = backgroundCanvas.getContext("2d");
const foreground = foregroundCanvas.getContext("2d");
backgroundCanvas.width = innerWidth;
backgroundCanvas.height = innerHeight;
foregroundCanvas.width = innerWidth;
foregroundCanvas.height = innerHeight;
let mouseButtonDown = false;
document.addEventListener('mousedown', () => mouseButtonDown = true);
document.addEventListener('mouseup', () => mouseButtonDown = false);
// Create objects to represent the current properties of the red and blue circle
const red = { x: 50, y: 50, radius: 20, color: "red" };
const blue = { x: 100, y: 100, radius: 10, color: "blue", angle: 0};
function drawCircle(ctx, {x, y, radius, color}) {
//--- Draw a circle to the specified canvas context, ctx = foreground or background
ctx.fillStyle = color;
ctx.beginPath();
ctx.arc(x, y, radius, 0, Math.PI*2);
ctx.closePath();
ctx.fill();
}
function animate(){
if (mouseButtonDown) {
red.x += (blue.x - red.x) / 10;
red.y += (blue.y - red.y) / 10;
} else {
blue.angle += 0.1;
}
blue.x = red.x + 35 * Math.sin(blue.angle);
blue.y = red.y + 35 * Math.cos(blue.angle);
drawCircle(background, red); // Draw the red circle in the background (without clearing the existing circles)
foreground.clearRect(0, 0, foregroundCanvas.width, foregroundCanvas.height); // Clear the foreground
drawCircle(foreground, blue); // Draw the blue circle on the foreground
requestAnimationFrame(animate);
}
animate();
Either way, it's convenient to abstract out the circle drawing code into a function or method, and to store the properties of the two circles in objects.
As #Blindman67's answer notes, there may be a performance cost of stacking 2 canvases, and if that is an issue you may want to try drawing the background offscreen then copying it to the on-screen canvas.
If you're not opposed to just building a particle class you can do it using them. In the snippet below I have a Circle class and a Particles class to creat what you are trying to achieve. I currently have the particles max at 500 but you can change it or delete that line all together if you ne er want them gone.
const canvas = document.getElementById("canvas");
const ctx = canvas.getContext("2d");
canvas.width = window.innerWidth;
canvas.height = window.innerHeight;
let mouseButtonDown = false;
//the array holding particles
let particles = [];
//the counter is only needed it you want to slow down how fast particles are being pushed and dispolayed
let counter = 0;
document.addEventListener("mousedown", () => (mouseButtonDown = true));
document.addEventListener("mouseup", () => (mouseButtonDown = false));
//ES6 constructor class
class Circle {
//sets the basic structor of the object
constructor(r, c) {
this.x = 100;
this.y = 100;
this.x2 = 50;
this.y2 = 50;
this.r = r; //will be assigned the argument passed in through the constructor by each instance created later
this.color = c; //same as above. This allows each instance to have different parameters.
this.angle = 0;
}
//this function creates the red circle
drawRed() {
ctx.fillStyle = this.color;
ctx.beginPath();
ctx.arc(this.x, this.y, this.r, 0, Math.PI * 2);
ctx.fill();
ctx.closePath();
}
//this function creates the blue circle
drawBlue() {
ctx.fillStyle = this.color;
ctx.beginPath();
ctx.arc(this.x + this.x2, this.y + this.y2, this.r, 0, Math.PI * 2);
ctx.fill();
ctx.closePath();
}
//this function is where we'll place parameter that change our object
update() {
//makes the blue circle rotate
this.x2 = 35 * Math.sin(this.angle);
this.y2 = 35 * Math.cos(this.angle);
//mouse action is same as your code
if (mouseButtonDown) {
this.x += this.x2 / 20;
this.y += this.y2 / 20;
} else {
this.angle += 0.1;
}
}
}
//When using this type of constructor class you have to create an instance of it by calling new Object. You can create as money as you want.
let blueCircle = new Circle(10, "blue"); //passing in the radius and color in to the constructor
let redCircle = new Circle(20, "red");
//another class for the particles
class Particles {
constructor() {
this.x = redCircle.x;
this.y = redCircle.y;
this.r = redCircle.r;
this.color = redCircle.color;
}
draw() {
ctx.fillStyle = this.color;
ctx.beginPath();
ctx.arc(this.x, this.y, this.r, 0, Math.PI * 2);
ctx.fill();
ctx.closePath();
}
}
//just wrapping all of the particle stuff into one function
function handleParticles() {
//while the mouse is held it will push particles
if (mouseButtonDown) {
particles.push(new Particles());
}
//this loops through the array and calls the draw() function for each particle
for (let i = 0; i < particles.length; i++) {
particles[i].draw();
}
//this keeps the array from getting too big.
if (particles.length > 500) {
particles.shift();
}
}
//wrap all functions into this one animate one and call requeatAnimationFrame
function animate() {
ctx.clearRect(0, 0, canvas.width, canvas.height);
handleParticles();
//These must be called for each instance created of the object
blueCircle.drawBlue();
blueCircle.update();
redCircle.drawRed();
redCircle.update();
requestAnimationFrame(animate);
}
animate();
#canvas1{
position: absolute;
top:0;
left: 0;
width: 100%;
height: 100%;
}
<canvas id="canvas"></canvas>
I'd also like to add you can change the rate that the particles are drawn by adding a counter variable and then limiting the draw like counter % 10 == 0
EXAMPLE
add global variable let counter = 0;
then in the handleParticles function add this
function handleParticles() {
counter++
if (mouseButtonDown && counter % 10 == 0) {
particles.push(new Particles());
}
for (let i = 0; i < particles.length; i++) {
particles[i].draw();
}
if (particles.length > 500) {
particles.shift();
}
}
Say we have a canvas:
<canvas id="one" width="100" height="200"></canvas>
var canvas = document.getElementById("one");
var context = canvas.getContext("2d");
var cw = canvas.width;
var ch = canvas.height;
// Sample graphic
context.beginPath();
context.rect(10, 10, 20, 50);
context.fillStyle = 'yellow';
context.fill();
context.lineWidth = 7;
context.strokeStyle = 'black';
context.stroke();
// create button
var button = document.getElementById("rotate");
button.onclick = function () {
// rotate the canvas 90 degrees each time the button is pressed
rotate();
}
var myImageData, rotating = false;
var rotate = function () {
if (!rotating) {
rotating = true;
// store current data to an image
myImageData = new Image();
myImageData.src = canvas.toDataURL();
myImageData.onload = function () {
// reset the canvas with new dimensions
canvas.width = ch;
canvas.height = cw;
cw = canvas.width;
ch = canvas.height;
context.save();
// translate and rotate
context.translate(cw, ch / cw);
context.rotate(Math.PI / 2);
// draw the previows image, now rotated
context.drawImage(myImageData, 0, 0);
context.restore();
// clear the temporary image
myImageData = null;
rotating = false;
}
}
}
And on a button click the canvas gets rotated -90 degrees anticlockwise (around the centre) and the dimensions of the canvas get also updated, so in a sense, it looks like this afterwards:
I want to rotate a canvas element to the anticlockwise rotation. I have used this code but it's not working as I want.
JavaScript has a built-in rotate() function for canvas context:
context.rotate( angle * Math.PI / 180);
The problem is that the rotation will only affect drawings made AFTER the rotation is done, which means you will need to:
Clear the canvas first: context.clearRect(0, 0, canvas.width, canvas.height);
Rotate the context context.rotate( 270 * Math.PI / 180);
Redraw the graphics
Thus, I recommend wrapping the graphics we want to draw in a function to make it easier to call after every rotation:
function drawGraphics() {
context.beginPath();
context.rect(10, 10, 20, 50);
context.fillStyle = 'yellow';
context.fill();
context.lineWidth = 7;
context.strokeStyle = 'black';
context.stroke();
}
I'm currently working on an HTML canvas example that fades out a line after it is drawn on a canvas, though my question is: how do I attain this while using a background image?
Live Demo (current code in a link at the bottom of this post)
CSS:
canvas {
background-image: url("https://zgab33vy595fw5zq-zippykid.netdna-ssl.com/wp-content/uploads/2018/02/PluralsightandSO.jpg");
background-size: 100% 100%;
}
JavaScript:
var canvas = document.getElementById("canvas"),
ctx = canvas.getContext("2d"),
painting = false,
lastX = 0,
lastY = 0;
canvas.width = canvas.height = 600;
canvas.onmousedown = function (e) {
if (!painting) {
painting = true;
} else {
painting = false;
}
lastX = e.pageX - this.offsetLeft;
lastY = e.pageY - this.offsetTop;
};
canvas.onmousemove = function (e) {
if (painting) {
mouseX = e.pageX - this.offsetLeft;
mouseY = e.pageY - this.offsetTop;
ctx.beginPath();
ctx.moveTo(lastX, lastY);
ctx.lineTo(mouseX, mouseY);
ctx.stroke();
lastX = mouseX;
lastY = mouseY;
}
}
function fadeOut() {
ctx.fillStyle = "rgba(255,255,255,0.3)";
ctx.fillRect(0, 0, canvas.width, canvas.height);
setTimeout(fadeOut,100);
}
fadeOut();
I realize fillRect is filling the entire canvas with the fillStyle so I assumed I could add the image to the canvas instead, but as you can tell, the animation is not the same by using this code instead:
var image = new Image();
image.src = "https://zgab33vy595fw5zq-zippykid.netdna-ssl.com/wp-content/uploads/2018/02/PluralsightandSO.jpg"
ctx.drawImage(image, 0, 0);
Current Live Demo (after changes described above) - it's too rigid and doesn't have as long of a tail
Any thoughts?
You're drawing a 30% transparent square to get the 'fading' effect:
ctx.fillStyle = "rgba(255,255,255,0.3)";
So you need to draw the PNG transparently as well
canvas.onmousemove = function (e) {
if (painting) {
// set line alpha to 1
ctx.globalAlpha = 1.0;
// your paint code here
}
}
function fadeOut() {
var image = new Image();
// set image alpha to 0.3
ctx.globalAlpha = 0.3;
// your draw image code here
}
By the way, you don't need to create the new image every time!
Example
You need to specify width and height:
ctx.drawImage(image, 0, 0, 600, 600);
I'm developing web app using canvas and I made three. canvas, canvas_panorama and canvas_image.
First one is something like main canvas, conteiner for the others. canvas_panorama is a background for canvas_image.
After canvas is right clicked, I'm computing angle to rotate canvas_image:
function getAngle( e, pw /*canvas*/ ){
var offset = pw.offset();
var center_x = (offset.left) + ($(pw).width() / 2);
var center_y = (offset.top) + ($(pw).height() / 2);
var mouse_x = e.pageX;
var mouse_y = e.pageY;
var radians = Math.atan2(mouse_x - center_x, mouse_y - center_y);
angle = radians;
}
After I have an angle I'm trying to rotate canvas_image like this:
function redraw(){
var p1 = ctx.transformedPoint(0,0);
var p2 = ctx.transformedPoint(canvas.width,canvas.height);
ctx.clearRect( p1.x, p1.y, p2.x-p1.x, p2.y-p1.y );
canvas_image_ctx.drawImage(image_img, 0, 0, 150, 150);
canvas_panorama_ctx.drawImage(panorama_img, 0, 0, 600, 300);
canvas_panorama_ctx.drawImage(canvas_image, 20, 20);
// rotate panorama_img around its center
// x = x + 0.5 * width
// y = y + 0.5 * height
canvas_panorama_ctx.translate(95, 95);
canvas_panorama_ctx.rotate(angle);
// translate to back
canvas_panorama_ctx.translate(-95, -95);
ctx.drawImage(canvas_panorama, 0, 0);
}
But this rotates both canvas_image and canvas_panorama. It should only rotate canvas_image
JSFiddle to show you my problem
I think you are confusing yourself with this idea of multiple canvases.
Once in the drawImage() method, every of your canvases are just images, and could be just one or even just plain shapes.
Transformation methods do apply to the canvas' context's matrix, and will have effect only if you do some drawing operations when they are set.
Note : To reset your context matrix, you can either use save(); and restore() methods which will also save all other properties of your context, so if you only need to reset the transform, then it's preferred to simply reset the transformation matrix to its default : ctx.setTransform(1,0,0,1,0,0).
Here is a simplified example to make things clearer :
var ctx = canvas.getContext('2d');
// a single shape, with the border of the context matrix
var drawRect = function(){
ctx.beginPath();
ctx.rect(10, 10, 50, 20);
ctx.fill();
ctx.stroke();
ctx.beginPath();
ctx.rect(0, 0, canvas.width, canvas.height);
ctx.stroke();
};
// set the color of our shapes
var gradient = ctx.createLinearGradient(0,0,70,0);
gradient.addColorStop(0,"green");
gradient.addColorStop(1,"yellow");
ctx.fillStyle = gradient;
// here comes the actual drawings
//we don't have modified the transform matrix yet
ctx.strokeStyle = "green";
drawRect();
// here we translate of 100px then we do rotate the context of 45deg
ctx.translate(100, 0)
ctx.rotate(Math.PI/4)
ctx.strokeStyle = "red";
drawRect();
// reset the matrix
ctx.setTransform(1,0,0,1,0,0);
// here we move of 150px to the right and 25px to the bottom
ctx.translate(150, 25)
ctx.strokeStyle = "blue";
drawRect();
// reset the matrix
ctx.setTransform(1,0,0,1,0,0);
<canvas id="canvas" width="500" height="200"></canvas>
In your code, you are setting the transformations on the canvas that does represent your image, and you do draw every of your canvases at each call.
What you want instead, is to set the transformation on the main canvas only, and draw the non-transformed image :
var main_ctx = canvas.getContext('2d');
var img_canvas = canvas.cloneNode();
var bg_canvas = canvas.cloneNode();
var angle = 0;
// draw on the main canvas, and only on the main canvas
var drawToMain = function(){
// first clear the canvas
main_ctx.clearRect(0,0,canvas.width, canvas.height);
// draw the background image
main_ctx.drawImage(bg_canvas, 0,0);
// do the transforms
main_ctx.translate(img_canvas.width/2, img_canvas.height/2);
main_ctx.rotate(angle);
main_ctx.translate(-img_canvas.width/2, -img_canvas.height/2);
// draw the img with the transforms applied
main_ctx.drawImage(img_canvas, 0,0);
// reset the transforms
main_ctx.setTransform(1,0,0,1,0,0);
};
// I changed the event to a simple onclick
canvas.onclick = function(e){
e.preventDefault();
angle+=Math.PI/8;
drawToMain();
}
// a dirty image loader
var init = function(){
var img = (this.src.indexOf('lena')>0);
var this_canvas = img ? img_canvas : bg_canvas;
this_canvas.width = this.width;
this_canvas.height = this.height;
this_canvas.getContext('2d').drawImage(this, 0,0);
if(!--toLoad){
drawToMain();
}
};
var toLoad = 2;
var img = new Image();
img.onload = init;
img.src = "http://pgmagick.readthedocs.org/en/latest/_images/lena_scale.jpg";
var bg = new Image();
bg.onload = init;
bg.src = 'http://www.fnordware.com/superpng/pnggradHDrgba.png';
<canvas id="canvas" width="500" height="300"></canvas>
I wrote some code that draws a semi-transparent rectangle over an image that you can draw with touch:
function drawRect() {
var canvas = document.getElementById('receipt');
var ctx = canvas.getContext('2d');
var drag = false;
var imageObj;
var rect = { };
var touch;
canvas.width = WIDTH;
canvas.height = HEIGHT;
function init() {
imageObj = new Image();
imageObj.src = 'img.jpg';
imageObj.onload = function() {
ctx.drawImage(imageObj, 0, 0);
};
canvas.addEventListener('touchstart', handleTouch, false);
canvas.addEventListener('touchmove', handleTouch, false);
canvas.addEventListener('touchleave', handleEnd, false);
canvas.addEventListener('touchend', handleEnd, false);
}
function handleTouch(event) {
if (event.targetTouches.length === 1) {
touch = event.targetTouches[0];
if (event.type == 'touchmove') {
if (drag) {
rect.w = touch.pageX - rect.startX;
rect.h = touch.pageY - rect.startY ;
draw();
}
} else {
rect.startX = touch.pageX;
rect.startY = touch.pageY;
drag = true;
}
}
}
function handleEnd(event) {
drag = false;
}
function draw() {
drawImageOnCanvas();
ctx.fillStyle = 'rgba(0, 100, 255, 0.2)';
ctx.fillRect(rect.startX, rect.startY, rect.w, rect.h);
}
function drawImageOnCanvas() {
ctx.clearRect(0, 0, canvas.width, canvas.height);
ctx.drawImage(imageObj, 0, 0);
}
init();
}
This works. But, now I'd like to have it so that I can capture each of the parts of the image that are in rectangles as separate images.
How can I pull this off? Because I have that redraw stuff, it deletes the previous rectangle, which makes this difficult.
A canvas can only save out the whole canvas as an image, so the trick is to create a temporary canvas the size of the region you want to save out.
One way could be to create a function which takes the image and a rectangle object as argument and returns a data-uri (or Blob) of the region:
function saveRegion(img, rect) {
var canvas = document.createElement("canvas"),
ctx = canvas.getContext("2d");
canvas.width = rect.w;
canvas.height = rect.h;
ctx.drawImage(img, rect.startX, rect.startY, rect.w, rect.h, 0, 0, rect.w, rect.h);
return canvas.toDataURL():
}
You can pass in the original image if don't want any graphics on top, or if you draw elements on top of it just pass in the original canvas element as image source. And of course, CORS-restrictions apply.