RemoveEventListener added with bind - Easy

In html5 canvases, when you add an event listener to something, it persists even if the movie clip is no longer on the stage. This causes two problems:

1) This takes up memory because the memory will hold on to the movie clip as long as it still has event listeners. Not just for movie clips on previous frames, but also if you are dynamically adding and removing clips - they will never go away, and your animation will gradually (or quickly) grind to a halt.

2) You an accidentally add multiple listeners to movie clips, e.g. when adding a mouseup event after a mousedown event, which you might do for dragging a movie clip. This could lead to contradictory or exaggerated responses, as well as taking up memory.

You can check if movieclips have event listeners before adding them with 'hasEventListener' (pass the event), and remove event listeners with 'removeEventListener' (pass the event, and the function).

This won't work if you have used the bind function to control the scope of the function. You have to pass the bound version that was originally given. If you are adding the function to multiple movieClips, creating a variable for each bound function could get messy. One way around this is to get each movieClip to store it's own bound functions.

See the example below:

------------------------------

1    this.tempFunction = function(e) {
2        console.log('click');
3        this.removeEventListener('click',this.ck);
4    }
5
6    this.btn.addEventListener('click',this.btn.ck = this.tempFunction.bind(this.btn));

------------------------------

You should find that the first time you click the button, a message 'click' comes up in the console (Ctrl+Shift+i in Chrome), and then it stops responding.

6: This line adds an event listener to the button (this.btn), that will respond to a click. The function called in response to the click is this.tempFunction, and bind is used so that the tempFunction works from the perspective (or scope) of the button. 

This part: "this.btn.ck = this.tempFunction.bind(this.btn)" simultaneously passes a function that should respond to the event and assigns it to a variable at the same time, and this variable (this.btn.ck) belongs to the button (this.btn).

1: This line declares the function that will respond to the click event of btn.

3: Remember that when this function is called, it will run from the scope of the btn, so when the event is removed, we can pass it 'this.ck', because from the btn scope, that is how to refer to the function that was referred to in line 6.

WARNING: removeEventListener won't tell you if it doesn't work, i.e., if it can't find the event you are trying to remove, so check the event has really been removed.

Getting the position of a character in a text object in CreateJS

 For those of us who are accustomed to Flash, if we wanted to know the exact x,y coordinates of a character in a textbox, using AS3, we could use the getCharBoundaries function, which if I remember correctly, would return a Rectangle object with the x, y, width and height... I think? The point is, there was a built-in method to do it.

Such a function does not exist in CreateJS (or I couldn't find one).

I was trying to make a little missing-words-type activity, and I wanted users to select the word they were dealing with (could this have been done in HTML with less bother? Probably, I dunno) - so I needed to make a button that would go behind the part of the textbox where the specific word was created.

When I saw that CreateJS did not have an equivalent of the getCharBoundaries function in AS3, I was ready to give up and try an alternative approach. But then I remembered that there are monospaced fonts such as Courier New, that we use in biosciences so that DNA and protein sequences are easy to read. If I used one of those (please skip to the code excerpt(s) below if you aren't interested in the evolution of this, I just really enjoyed this little problem-solving journey, personally), I reasoned, then as long as I knew where the text started, I might be able to figure out the location of each letter, though working out where the line breaks were would be a pain.

On the stage, I manually measured the width of each letter (I realised later a better way of doing this, see below) at 21.6px in the font and size I was using. I used this letter width and the index of the character in the text to position a movieclip relative to the text area.

There were two empty movieclips on stage, the one on top (loadIt) was for loading the dynamic text, the one underneath (loadItBtn) was for loading the buttons.

It went something like this (just focusing on how to deal with a single line first):

tx = new createjs.Text('Hello _____.''36px Courier New','black');
this.loadIt.addChild(tx);

this.letterWidth = 21.6;//I measured this on the stage manually...blech
for(i=6; i<11; i++) {
	btn = new lib.btnMC();//this is just a rectangle made to match the height of the line, and the width of the letters.
	btn.x = i * this.letterWidth;
	this.loadItBtn.addChild(btn);
}

So, some useful functions: tx.getMeasuredWidth() could tell me how wide the text area was, as I was using short paragraphs. I would then use tx.lineWidth, to set how wide I wanted the text area to be, which would cause word wrapping. I would then use tx.getMeasuredHeight() to find out the height of the text area so I could position it on the canvas (you can use a function to get or set your lineHeight).

However, what if the text ends up with multiple lines? Enter (drumroll) getMetrics()!

This function will give you an array, called 'lines', which tells you what text is in each line! So I wrote a loop to check whether the index of the character in the text could be found in the line of the array. Then adjusted the y coordinate of the buttons I was adding based on that. Then I had to remember to figure out what the index of my character was in the line of the text, and use that as the x coordinate.

I won't go into all the details, but if you use console.log(text_name.getMetrics());, you can see the output for yourself in the console.

ISSUES:

1) Using scaleX to size an mc to match the word width did not work for me, it doesn't seem to be very accurate. I dealt with this by making a movieclip that was the perfect size and adding it multiple times, but a better solution would have been to draw a rectangle of the correct size.

2) Measuring the letter width manually is a pain, but manageable because you only need to do it once, but then you have to stick to that font size. I realised later that you could write a function to create a text area, add some text and measure the width to calculate the width of a single letter. Then I realised that actually, you could do this for any font! You'd have to make sure you measured every type of letter you were planning to use, and punctuation, but you could easily create an object that stores the widths of the characters you are using. Easily-ish. But basically, it means that you could work out the position of letters in text areas, and such a function would require some alteration if the text was left-, centre- or right-justified, and it would probably only shorten your life-span by a few years... But, yeah, doable.

You can see an example of what I was aiming for here: Missing words exercise

Have a nice day!

How to: Input text box in Animate CC canvas, version 2

This method will allow you to put input text boxes within a particular location on your canvas, using co-ordinates etc that make sense, that will change if the page is resized.

By the end of it, you should be able to create an input textbox form within your canvas, just by calling a function like this:

createTextBox(250, 76.05, 315.1, 16);//width, xpos, ypos, fontsize

1) Put the canvas in a div, so that you can easily add input boxes just after the canvas.

 <div id='canvasHolder'>
<canvas id="canvas" width="800" height="450" style="display: block; background-color:rgba(0, 0, 0, 1.00)"></canvas>
</div>

2) My canvas is set to resize with the window, under Publish Settings, I chose: Center stage, Both; Make responsive, Both; Scale to fill visible area, Fit in view. So this code works for those settings, but seeing this might help if you are using something different.

So the canvas html contains a function that automatically resizes the canvas, and you will need some of those variables to figure out how big your textbox needs to be. So you need to give them a global scope. In the javascript,

var pRatio, xRatio, yRatio, sRatio;

Also, create an array that will hold all your text boxes so that you can resize them later.

var txtBoxes = new Array();

You also need to go into the resizing function and remove the 'var' so that they are global variables. In the resizeCanvas function, change:

var pRatio = window.devicePixelRatio || 1, xRatio=iw/w, yRatio=ih/h, sRatio=1;

to

pRatio = window.devicePixelRatio || 1, xRatio=iw/w, yRatio=ih/h, sRatio=1;

3) Now here is the function that will create the input text box:

function createTextBox(wid, xpos, ypos, fsize) {
var node = document.createElement("INPUT");
node.type = 'text';
node.style = "position: fixed; font-family:'Gill Sans MT'; border:none; border-bottom-style:solid; border-bottom-width:1px";
txtBoxes.push({'ele':node,'wid':wid,'xpos':xpos,'ypos':ypos,'fsize':fsize});
sizeBox(txtBoxes[txtBoxes.length-1]);
document.getElementById('canvasHolder').appendChild(node);
}

The first line of this function creates an input 'node', the second specifies it as a text input, although this is the default, so not strictly necessary.

Setting the css position to fixed works for me, but some others might work too, this isn't my strong point. Setting the font and border was a personal choice.

It adds an object containing the node and all its sizing properties to the txtBoxes array, so that this info can be used when resizing later.

It calls the sizeBox function with the object - this will position the object appropriately.

Then it adds the node to the canvasHolder element (where the canvas is).

4) Here is the sizeBox function:

function sizeBox(obj) {
var node = obj['ele'];
node.style.width =  Math.round((canvas.width/pRatio)*(obj['wid']/lib.properties.width)) + "px";
node.style.fontSize = Math.round(sRatio*obj['fsize'])+"px";
if(xRatio < yRatio) {
node.style.left = Math.round((canvas.width/pRatio)*(obj['xpos']/lib.properties.width))+"px";
node.style.top = Math.round((0.5*(window.innerHeight-(canvas.height/pRatio)))+((canvas.height/pRatio)*(obj['ypos']/lib.properties.height)))+"px";
}else{
node.style.left = Math.round((0.5*(window.innerWidth - (canvas.width/pRatio))) + ((canvas.width/pRatio) * (obj['xpos']/lib.properties.width))) + "px";
node.style.top = Math.round((canvas.height/pRatio)*(obj['ypos']/lib.properties.height)) + "px";
}
}

The size properties of the canvas is complex and I don't fully understand how it works, but I just kept fiddling around with things until it sort of made sense. Basically, you can set the width relative to the canvas' actual size, same for the font.

Its similar for the x and y position, but one will need to be offset because the ratio of the width and the height may not be equal.

5) Finally, alter the resizeCanvas function so that the textboxes are automatically altered.

After this line:

lastW = iw; lastH = ih; lastS = sRatio;

Add this:

for(i=0; i sizeBox(txtBoxes[i]);
}

It goes through each of the txtBoxes and resizes them, as the window is being resized.

6) Now you can create them from your canvas, like so:

createTextBox(250, 76.05, 315.1, 16);//width, xpos, ypos, fontsize


Just a few notes about this method, the alignment is 100% flawless, I think because everything has to be rounded off to pixels, but it is pretty good.
Also, how to access the value inside? You can't assign them an id, and use document.getElementById, because the node is dynamically added. But you can access the txtBoxes array and get the value of the node.

How to: Input text box in Animate CC Canvas, version 1

Am getting more used to animate CC canvas, and was quite happy until I found that you couldn't do input text boxes, at least not directly. But I suppose the nice thing about these difficulties is the smug feeling of self-satisfaction you get when you overcome it (entirely undeserved, with all the online help I got), anyway!

The strategy is basically to make an invisible text box that holds on to text when you type and passes it to the text box on your canvas.

In terms of making the text box invisible, you can't use display: none, as textboxes can't get the focus if they are not displayed, or hidden.

If you use opacity:0, then you won't be able to see the textbox, but it will still take up space, and someone could click on it by accident, possibly leading to upredictable results.

So I'm opting to use opacity, set its width to 0, and setting it to be off of the screen as well!

1)
Here is the html for the textbox:

<input type='text' id='testInput' oninput="sendInput();" style="position:absolute; z-index:-1; opacity:0; width:0; top:-30px">

If this is the very first element in the body tag, then it should be above the visible window, ie offstage. If you have other elements around your canvas you may need to fiddle with this, but I doubt it as the textbox doesn't need to be in any particular div to work.


2)
In the script tag in the head, you need to create variables for your input textbox, and whatever the receiving textbox might be, something like:

<script>
var txtTarget, txtInput;

And in the init function called by the loading body tag:

<body onLoad="init()">

You need to assign the input textbox to the txtInput variable:

function init() {
     txtInput = document.getElementById('testInput');
}

3)
As for the sendInput function, that goes in the script section:

function sendInput() {
 txtTarget.text = txtInput.value;
}

This sends the content of the input text box to the text box in the canvas element.

4)
In the animate CC flash file, you need to create a dynamic text box (call it textReceiver), and a button behind it (called txtbtn), you could save time by making it look like an input textbox.

The code for this is:

this.tstbtn.on('click', focusText.bind(this));
function focusText(e) {
 //txtInput.value = this.testReceive.text = '';
 txtInput.focus();
 txtTarget = this.testReceive;
}

This function means that when someone clicks on the button, it is as though they have clicked on the input textbox. It also sets the html page to recognise the canvas's text box as the text receiver.


In theory, you could create multiple textboxes (though not on the same frame in animate CC), and put a button behind each one, so that when it is clicked, it registers its textbox as txtTarget. In that case, you need to use the line that is commented out, so that the text from one textbox doesn't get copied over to another.

The downsides to this method, which occurred to me as I was writing the post, are that the user doesn't get a cursor, they can't highlight the text, and they can't easily edit the content, since they can't click part way through the text.

So I may go back to the drawing board on this, to get the full input text box functionality.

Flash CC Pro/CreateJS: nextFrame

This function doesn't exist in Flash's HTML5 canvas, but the below will work instead.

this.forw.addEventListener("click", goNext.bind(this));
function goNext() {
this.gotoAndStop(this.currentFrame + 1);
}

Flash CC Pro/CreateJS

Another strange thing when working with Flash's HTML5 Canvas is that variables cease to exist when you move to the next frame.

I am used to having a root timeline that I can store info on that all other movieclips might need, but this won't work if they disappear between one frame and the next.

My current adjustment is to move everything down by one timeline. So I have a single frame on the uppermost timeline, and the movie clip sitting on that contains several frames, as I would normally lay things out. If something needs to be stored while moving to the next frame, then I just store it on the parent timeline.

for this function that I wanted all movie clips to be able to access:

this.giveDimensions = function (mc) {
mc.height = mc.nominalBounds.height * mc.scaleY;
mc.width = mc.nominalBounds.width * mc.scaleX;
}

In the timeline of the movieclip:

this.parent.giveDimensions(this.work_btn);

Flash CC Pro/CreateJS: Width and height of movieclips

I am currently trying out Flash CC Pro and the HTML5 Canvas publishing option.

One odd thing is not being able to access the height and width of the movie clips.

With the help of a stackoverflow post (http://stackoverflow.com/questions/29768424/how-to-get-width-height-of-a-movie-clip-in-createjs), I came across 'nominal bounds'. Couldn't find it in any of the documentation, it doesn't appear as one of the coding hints in the flash program either. Oh well.

I found that movieClip.nominalBounds gave me back an object that contained the original x, y, width and height of the unaltered movie clip. But on the stage, mine was a bit stretch and squashed, so I used

movieClip.nominalBounds.height * movieClip.scaleY

and this gave me the actual height. Ditto for the width.

Apparently the various bounds properties and functions vary depending on the exact type of object, but this worked for my movie clip.

How to randomly generate -1 or +1, in one line.

A hobby of mine is avoiding if statements, with mathematical formula that return the right number based on other conditions.

It is easy to generate a 0 or a 1 randomly, simply use Math.round(Math.random());

That will work in a number of languages, or something similar.

For a random -1 or +1, try this:

var randNum = -1 + (2*Math.round(Math.random));

If the random number generated by Math.random() is 0.5 or above, then the answer will be 1. If the random number generated is less than 0.5, the answer will be 1.

Enjoy!

How to make a RevoluteJoint in Box2D/CreateJS

There are plenty of posts on how to get box2d and createjs working, Lee Brimelow has a great youtbe video that covers how to get started very nicely.

But when I tried to make a revolute joint, I made a silly mistake that took me hours to find the source of. I'm a bit of a newbie, but I didn't find the problem to my solution online, so I thought I'd post it in case it happens to anyone else. And then maybe they can go to sleep before 5 in the morning...

Does this sound familiar?

"I can make the two objects, but when I try to create the joint, I get an 'Uncaught TypeError: undefined is not a function' message.'

Well, first, since I was working from the Lee Brimelow example as a starting point, I hadn't made sure I had setup the shortcut to the actual RevoluteJoint parts of box2d.

eg.
var box2d = {
b2Vec2 : Box2D.Common.Math.b2Vec2,
b2BodyDef : Box2D.Dynamics.b2BodyDef,
b2Body: Box2D.Dynamics.b2Body,
b2FixtureDef: Box2D.Dynamics.b2FixtureDef,
b2Fixture: Box2D.Dynamics.b2Fixture,
b2RevoluteJointDef: Box2D.Dynamics.Joints.b2RevoluteJointDef,
b2RevoluteJoint: Box2D.Dynamics.Joints.b2RevoluteJoint,
b2World: Box2D.Dynamics.b2World,
b2MassData: Box2D.Collision.Shapes.b2MassData,
b2PolygonShape: Box2D.Collision.Shapes.b2PolygonShape,
b2CircleShape: Box2D.Collision.Shapes.b2CircleShape,
b2DebugDraw: Box2D.Dynamics.b2DebugDraw
};

Still wouldn't work though. So here was the problem, my makeJoint function looked like this:
function makeJoint() {
var jDef = new box2d.b2RevoluteJointDef();
jDef.bodyB = c;
jDef.bodyA = a;
jDef.collideConnected = false;
jDef.localAnchorB = new box2d.b2Vec2(0/SCALE,80/SCALE);
jDef.localAnchorA = new box2d.b2Vec2(0,0);
j = world.CreateJoint(jDef);
}

where a (I thought) was the anchor body, and c was the other body I wanted to attach to it.

Unfortunately, a and c were actually the fixtures of those bodies, instead of the bodies themselves.

So in my function that made the anchor body, it said:
a = world.CreateBody(anchorBodyDef).CreateFixture(anchorFixtureDef) ;

So I changed it to:
a = world.CreateBody(anchorBodyDef);
a.CreateFixture(anchorFixtureDef);

Hope that helps.

CreateJS: Change Background Colour of Canvas

Very simple, using CSS

Create JS: EaselJS: Getting Started

What is CreateJS?
For me, CreateJS will be a way to take my Flash skills and use them in a way that will work with HTML5, and on more devices than the Flash Player is currently running on.

First things first - You need to know HTML and JavaScript. A great place to learn this is at W3Schools, don't worry, it's free!

I'd start with learning HTML, then HTML5, then JavaScript.

CreateJS appears to be a JavaScript 'library' (a collection of JavaScript functions that someone has written, that makes really complex things a lot easier), which also writes a lot like actionscript 3 (Flash programming language). So if you know JavaScript, or ActionScript 3, then a lot of the logic behind what you are doing comes very quickly.

Second things second - You need to download the library.
It looks like you can actually use the online versions of the libraries in your scripts, but I think your fields would have to be online too for that to work.

For writing and testing script on your hard-drive, the files need to be on your hard-drive too (I tried accessing the online one, it wouldn't work. I'm sure some people knew this was obvious, but I had to check.).

The website for downloading them is here. And you will have to unzip the folder to access the files inside.

Once inside, open the 'tutorials' folder, then the 'Getting Started' folder, and then the 'index.html' file. This will take you through the basics of how to use the library as a script source, making a canvas tag, and if I remember correctly, drawing a red circle.

For now, I will be going through these tutorials, so anything I post will just be a re-organisation of what I learnt there, into a format that makes more sense to me.

But I just thought I'd post this, because I spent a long time trying to find free tutorials on line, only to find that they were available within the download itself!

Warning: The tutorial may have typos in the code (e.g. Window, instead of window), but the source code for the examples works, so use that as a guide.