I've tried a few different ways that I have seen on here, but I can't quite get my image to move. Whenever I try adapting code for arrow key presses, it just seems to make my canvas shrink and my player model (spaceperson) disappear.
here is the "drawing board" I keep returning to, and what I have so far.
// Get the canvas and context
var canvas = document.getElementById("space");
var ctx = canvas.getContext("2d");
canvas.width = 1920;
canvas.height = 700;
// Create the image object
var spaceperson = new Image();
// Add onload event handler
spaceperson.onload = function () {
// Done loading, now we can use the image
ctx.drawImage(spaceperson, 280, 300);
};
// artwork by Harrison Marley (using make8bitart.com)
spaceperson.src = "http://i.imgur.com/Eh9Dpq2.png";`
I am quite new to javascript, and I am just trying to work out how I can move the specperson image using arrow keys. I was trying to make a class for space person to access their x,y values, but I can't seem to draw the image without using .onload
here a more complete example:
//just a utility
function image(url, callback){
var img = new Image();
if(typeof callback === "function"){
img.onload = function(){
//just to ensure that the callback is executed async
setTimeout(function(){ callback(img, url) }, 0)
}
}
img.src = url;
return img;
}
//a utility to keep a value constrained between a min and a max
function clamp(v, min, max){
return v > min? v < max? v: max: min;
}
//returns a function that can be called with a keyCode or one of the known aliases
//and returns true||false wether the button is down
var isKeyDown = (function(aliases){
for(var i=256, keyDown=Array(i); i--; )keyDown[i]=false;
var handler = function(e){
keyDown[e.keyCode] = e.type === "keydown";
e.preventDefault(); //scrolling; if you have to suppress it
};
addEventListener("keydown", handler, false);
addEventListener("keyup", handler, false);
return function(key){
return(true === keyDown[ key in aliases? aliases[ key ]: key ])
}
})({
//some aliases, to be extended
up: 38,
down: 40,
left: 37,
right: 39
});
// Get the canvas and context
var canvas = document.getElementById("space");
canvas.width = 1920;
canvas.height = 700;
var ctx = canvas.getContext("2d");
//the acutal image is just a little-part of what defines your figue
var spaceperson = {
image: image("//i.imgur.com/Eh9Dpq2.png", function(img){
spaceperson.width = img.naturalWidth;
spaceperson.height = img.naturalHeight;
//start the rendering by calling update
update();
}),
//position
x: 60, y: 310,
width: 0, height: 0,
speed: 200 // 200px/s
};
var lastCall = 0; //to calculate the (real) time between two update-calls
//the render-fucntion
function update(){
//taking account for (sometimes changing) framerates
var now = Date.now(), time = lastCall|0 && (now-lastCall)/1000;
lastCall = now;
requestAnimationFrame(update);
var sp = spaceperson,
speed = sp.speed;
//checking the pressed buttons and calculates the direction
//two opposite buttons cancel out each other, like left and right
var dx = (isKeyDown('right') - isKeyDown('left')) * time,
dy = (isKeyDown('down') - isKeyDown('up')) * time;
//fix the speed for diagonals
if(dx && dy) speed *= 0.7071067811865475; // * 1 / Math.sqrt(2)
if(dx) { //there is some movement on the x-axes
sp.x = clamp(
//calculate the new x-Position
//currentPos + direction * speed
sp.x + dx * sp.speed,
//restraining the result to the bounds of the map
0, canvas.width - sp.width
);
}
//same for y
if(dy) sp.y = clamp(sp.y + dy * sp.speed, 0, canvas.height - sp.height);
ctx.clearRect(0, 0, canvas.width, canvas.height);
ctx.drawImage(sp.image, sp.x, sp.y);
}
Edit:
A quick question (I hope); if I was to later add other objects, would I check for collisions in update()?
This is still just a very basic example. The main purpose of the update()
-function should be to work as the main event-loop.
To trigger all Events that have to happen each frame in the order they have to happen.
var lastCall = 0;
function update(){
//I always want a next frame
requestAnimationFrame(update);
//handle timing
var now = Date.now(),
//time since the last call in seconds
//cause usually it's easier for us to think in
//tems like 50px/s than 0.05px/ms or 0.8333px/frame
time = lastCall|0 && (now-lastCall) / 1000;
lastCall = now;
movePlayer(time);
moveEnemies(time);
moveBullets(time);
collisionDetection();
render();
}
function render(){
ctx.clear(0, 0, canvas.width, canvas.height);
drawBackground(ctx);
for(var i=0; i<enemies.length; ++i)
enemies[i].render(ctx);
player.render(ctx);
}
Not saying that you have to implement all these functions now, but to give you an idea of a possible structure.
Don't be scared to break big tasks (functions) up into subtasks.
And it might make sense to give each enemy a move()
-function so you can implement different movement-patterns per enemy,
or you say that the pattern is (and will be) all the same for each enemy, parameterized at the best, then you can handle that in a loop.
Same thing for rendering, as I'm showing in the last part of code.