- 1.1. The canvas Element
- 1.2. Canvas Contexts
- 1.3. Canonical Examples in This Book
- 1.4. Getting Started
- 1.5. Fundamental Drawing Operations
- 1.6. Event Handling
- 1.7. Saving and Restoring the Drawing Surface
- 1.8. Using HTML Elements in a Canvas
- 1.9. Printing a Canvas
- 1.10. Offscreen Canvases
- 1.11. A Brief Math Primer
- 1.12. Conclusion
1.8. Using HTML Elements in a Canvas
Canvas is arguably the coolest feature of HTML5, but when you use it to implement web applications, you will rarely use it alone. Most of the time you will combine one or more canvases with other HTML controls so that your users can provide input or otherwise control the application.
To combine other HTML controls with your canvases, you may first be inclined to embed those controls inside your canvas elements, but that won’t work, because anything you put in the body of a canvas element is displayed by the browser only if the browser does not support the canvas element.
Because browsers will display either a canvas element or HTML controls that you put inside that element, but not both, you must place your controls outside of your canvas elements.
To make it appear as though HTML controls are inside a canvas, you can use CSS to place the controls above the canvas. The application shown in Figure 1.16 illustrates that effect.
Figure 1.16. HTML elements above a canvas
The application shown in Figure 1.16 animates 100 balls and provides a link to start and stop the animation. That link resides in a DIV element that is partially transparent and floats above the canvas. We refer to that DIV as a glass pane because it appears to be a pane of glass floating above the canvas.
The HTML for the application shown in Figure 1.16 is listed in Example 1.8.
Example 1.8. HTML controls in a canvas: HTML
<!DOCTYPE
html
><html>
<head>
<title>
Bouncing Balls</title>
<style>
body{
background:
#dddddd
;}
#canvas
{
margin-left:
10px
;margin-top:
10px
;background:
#ffffff
;border:
thin solid
#aaaaaa
;}
#glasspane
{
position:
absolute
;left:
50px
;top:
50px
;padding:
0px 20px 10px 10px
;background:
rgba
(0
,0
,0
,0.3
);border:
thin solid rgba
(0
,0
,0
,0.6
);color:#
#eeeeee
;font-family:
Droid Sans
,Arial
,Helvetica
,sans-serif
;font-size:
12px
;cursor:
pointer
;-webkit-box-shadow:
rgba
(0
,0
,0
,0.5
)5px 5px 20px
;-moz-box-shadow:
rgba
(0
,0
,0
,0.5
)5px 5px 20px
;box-shadow:
rgba
(0
,0
,0
,0.5
)5px 5px 20px
;}
#glasspane
h2{
font-weight:
normal
;}
#glasspane .title
{
font-size:
2em
;color:#
rgba
(255
,255
,0
,0.8
);}
#glasspane
a:
hover{
color:#
yellow
;}
#glasspane
a{
text-decoration:
none
;color:#
#cccccc
;font-size:
3.5em
;}
#glasspane
p{
margin:
10px
;color:#
rgba
(65
,65
,220
,1.0
);font-size:
12pt
;font-family:
Palatino
,Arial
,Helvetica
,sans-serif
;}
</style>
</head>
<body>
<div
id
=
'glasspane'
>
<h2
class
=
'title'
>
Bouncing Balls</h2>
<p>
One hundred balls bouncing</p>
<a
id
=
'startButton'
>
Start</a>
</div>
<canvas
id
=
'canvas'
width
=
'750'
height
=
'500'
>
Canvas not supported</canvas>
<script
src
=
'example.js'
></script>
</body>
</html>
The HTML shown in Example 1.8 uses CSS absolute positioning to make the glass pane appear above the canvas, like this:
#canvas
{
margin-left:
10px
;margin-top:
10px
;background:
#ffffff
;border:
thin solid
#aaaaaa
;}
#glasspane
{
position:
absolute
;left:
50px
;top:
50px
; ...}
The preceding CSS uses relative positioning for the canvas, which is the default for the position CSS property, whereas it specifies absolute positioning for the glass pane. The CSS specification states that elements with absolute positioning are drawn on top of elements with relative positioning, which is why the glass pane appears above the canvas in Figure 1.16.
If you also change the canvas’s positioning to absolute, then the canvas will appear on top of the glass pane, and you won’t see the glass pane because the canvas’s background is not transparent. In that case, the glass pane is underneath the canvas because the canvas element comes after the glass pane’s DIV element. If you switch the order of those elements, then the glass pane will once again appear above the canvas.
So, you have two options to position the glass pane above the canvas: Use relative positioning for the canvas and absolute positioning for the glass pane; or use either relative or absolute positioning for both elements and declare the glass pane’s DIV after the canvas element.
A third option is to use either relative or absolute positioning for both elements and manipulate their z-index CSS property. The browser draws elements with a higher z-index above elements with a lower z-index.
In addition to placing HTML controls where you want them to appear, you also need to obtain references to those elements in your JavaScript so that you can access and manipulate their values.
The application shown in Figure 1.16 obtains references to the glass pane and the button that controls the animation and adds event handlers to them, like this:
var
context=
document.
getElementById
(
'canvas'
)
.
getContext
(
'2d'
),
startButton=
document.
getElementById
(
'startButton'
),
glasspane=
document.
getElementById
(
'glasspane'
),
paused=
false
,
... startButton.
onclick=
function
(
e)
{
e.
preventDefault
();
paused= !
paused;
startButton.
innerText=
paused?
'Start'
:
'Stop'
;
}
;
... glasspane.
onmousedown=
function
(
e)
{
e.
preventDefault
();
}
;
The preceding JavaScript adds an onclick handler to the button that starts or pauses the animation based on the current state of the application, and adds an onmousedown event handler to the glass pane to prevent the browser from its default reaction to that mouse click. The onmousedown handler prevents the browser from reacting to the event to avoid inadvertent selections.
1.8.1. Invisible HTML Elements
In the preceding section you saw how to combine static HTML controls with a canvas. In this section we explore a more advanced use of HTML controls that involves dynamically modifying the size of a DIV as the user drags the mouse.
Figure 1.17 shows an application that uses a technique known as rubberbanding to select a region of a canvas. That canvas initially displays an image, and when you select a region of that image, the application reacts by zooming into the region that you selected.
Figure 1.17. Implementing rubber bands with a DIV
First, let’s take a look at the HTML for the application, which is listed in Example 1.9.
Example 1.9. Rubber band with a floating DIV
<!DOCTYPE
html
><html>
<head>
<title>
Rubber bands with layered elements</title>
<style>
body{
background:
rgba
(100
,145
,250
,0.3
);}
#canvas
{
margin-left:
20px
;margin-right:
0
;margin-bottom:
20px
;border:
thin solid
#aaaaaa
;cursor:
crosshair
;padding:
0
;}
#controls
{
margin:
20px 0px 20px 20px
;}
#rubberbandDiv
{
position:
absolute
;border:
3px solid blue
;cursor:
crosshair
;display:
none
;}
</style>
</head>
<body>
<div
id
=
'controls'
>
<input
type
=
'button'
id
=
'resetButton'
value
=
'Reset'
/>
</div>
<div
id
=
'rubberbandDiv'
></div>
<canvas
id
=
'canvas'
width
=
'800'
height
=
'520'
>
Canvas not supported</canvas>
<script
src
=
'example.js'
></script>
</body>
</html>
The HTML uses a DIV that contains a button. If you click that button, the application draws the entire image as it is displayed when the application starts.
The application uses a second DIV for the rubber band. That DIV is empty, and its CSS display attribute is set to none, which makes it initially invisible. When you start dragging the mouse, the application makes that second DIV visible, which shows the DIV’s border. As you continue dragging the mouse, the application continuously resizes the DIV to produce the illusion of a rubber band, as shown in Figure 1.17.
The JavaScript for the application shown in Figure 1.17 is listed in Example 1.10.
Example 1.10. Rubber bands with a DIV
var
canvas=
document.
getElementById
(
'canvas'
),
context=
canvas.
getContext
(
'2d'
),
rubberbandDiv=
document.
getElementById
(
'rubberbandDiv'
),
resetButton=
document.
getElementById
(
'resetButton'
),
image=
new
Image
(),
mousedown=
{}
,
rubberbandRectangle=
{}
,
dragging=
false
;
// Functions..........................................................function
rubberbandStart
(
x,
y)
{
mousedown.
x=
x;
mousedown.
y=
y;
rubberbandRectangle.
left=
mousedown.
x;
rubberbandRectangle.
top=
mousedown.
y;
moveRubberbandDiv
();
showRubberbandDiv
();
dragging=
true
;
}
function
rubberbandStretch
(
x,
y)
{
rubberbandRectangle.
left=
x<
mousedown.
x?
x:
mousedown.
x;
rubberbandRectangle.
top=
y<
mousedown.
y?
y:
mousedown.
y;
rubberbandRectangle.
width=
Math.
abs
(
x-
mousedown.
x),
rubberbandRectangle.
height=
Math.
abs
(
y-
mousedown.
y);
moveRubberbandDiv
();
resizeRubberbandDiv
();
}
function
rubberbandEnd
()
{
var
bbox=
canvas.
getBoundingClientRect
();
try
{
context.
drawImage
(
canvas,
rubberbandRectangle.
left-
bbox.
left,
rubberbandRectangle.
top-
bbox.
top,
rubberbandRectangle.
width,
rubberbandRectangle.
height,
0
,
0
,
canvas.
width,
canvas.
height);
}
catch
(
e)
{
// Suppress error message when mouse is released // outside the canvas}
resetRubberbandRectangle
();
rubberbandDiv.
style.
width=
0
;
rubberbandDiv.
style.
height=
0
;
hideRubberbandDiv
();
dragging=
false
;
}
function
moveRubberbandDiv
()
{
rubberbandDiv.
style.
top=
rubberbandRectangle.
top+
'px'
;
rubberbandDiv.
style.
left=
rubberbandRectangle.
left+
'px'
;
}
function
resizeRubberbandDiv
()
{
rubberbandDiv.
style.
width=
rubberbandRectangle.
width+
'px'
;
rubberbandDiv.
style.
height=
rubberbandRectangle.
height+
'px'
;
}
function
showRubberbandDiv
()
{
rubberbandDiv.
style.
display=
'inline'
;
}
function
hideRubberbandDiv
()
{
rubberbandDiv.
style.
display=
'none'
;
}
function
resetRubberbandRectangle
()
{
rubberbandRectangle=
{
top:
0
,
left:
0
,
width:
0
,
height:
0
}
;
}
// Event handlers..................................................... canvas.onmousedown=
function
(
e)
{
var
x=
e.
clientX,
y=
e.
clientY;
e.
preventDefault
();
rubberbandStart
(
x,
y);
}
;
window.
onmousemove=
function
(
e)
{
var
x=
e.
clientX,
y=
e.
clientY;
e.
preventDefault
();
if
(
dragging)
{
rubberbandStretch
(
x,
y)
;
}
}
;
window.
onmouseup=
function
(
e)
{
e.
preventDefault
();
rubberbandEnd
();
}
;
image.
onload=
function
()
{
context.
drawImage
(
image,
0
,
0
,
canvas.
width,
canvas.
height);
}
;
resetButton.
onclick=
function
(
e)
{
context.
clearRect
(
0
,
0
,
context.
canvas.
width,
context.
canvas.
height);
context.
drawImage
(
image,
0
,
0
,
canvas.
width,
canvas.
height);
}
;
// Initialization..................................................... image.
src=
'curved-road.png'
;
Again, we’re getting ahead of ourselves a little bit by using the drawImage() method to both draw and zoom in on the image. In Section 4.1, “Drawing Images,” on p. 254, we will look closely at that method, and we will also see an alternative way to implement rubber bands that involves manipulating the image’s pixels to draw the rubber band itself.
For now, however, our focus is on the rubberband DIV and how the code manipulates that DIV as the user drags the mouse.
The onmousedown event handler for the canvas invokes the rubberbandStart() method, which moves the DIV’s upper left-hand corner to the mouse down location and makes the DIV visible. Because the rubberband DIV’s CSS position attribute is absolute, the coordinates for the DIV’s upper left-hand corner must be specified in window coordinates, and not as coordinates relative to the canvas.
If the user is dragging the mouse, the onmousemove event handler invokes rubberbandStretch(), which moves and resizes the rubberband DIV.
When the user releases the mouse, the onmouseup event handler invokes rubberbandEnd(), which draws the scaled image and shrinks and hides the rubberband DIV.
Finally, notice that all three mouse event handlers invoke preventDefault() on the event object they are passed. As discussed in Section 1.6.1.1, “Translating Mouse Coordinates to Canvas Coordinates,” on p. 26, that call prevents the browser from reacting to the mouse events. If you remove those calls to preventDefault(), the browser will try to select elements on the page, which produces undesired effects if the user drags the mouse outside of the canvas.