path: root/doc/src/webkit/guide/chapter_canvas.qdoc

/****************************************************************************
**
** Copyright (C) 2011 Nokia Corporation and/or its subsidiary(-ies).
** All rights reserved.
** Contact: Nokia Corporation (qt-info@nokia.com)
**
** This file is part of the Qt WebKit documentation.
**
** $QT_BEGIN_LICENSE:BSD$
** You may use this file under the terms of the BSD license as follows:
**
** "Redistribution and use in source and binary forms, with or without
** modification, are permitted provided that the following conditions are
** met:
**   * Redistributions of source code must retain the above copyright
**     notice, this list of conditions and the following disclaimer.
**   * Redistributions in binary form must reproduce the above copyright
**     notice, this list of conditions and the following disclaimer in
**     the documentation and/or other materials provided with the
**     distribution.
**   * Neither the name of Nokia Corporation and its Subsidiary(-ies) nor
**     the names of its contributors may be used to endorse or promote
**     products derived from this software without specific prior written
**     permission.
**
** THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
** "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
** LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
** A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
** OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
** SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
** LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
** DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
** THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
** (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
** OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE."
** $QT_END_LICENSE$
**
****************************************************************************/
/*!

\page qtwebkit-guide-canvas.html
\title Canvas Graphics (BETA)

\chapter Canvas Graphics

HTML5\'s Canvas API enables you to draw within a Web page or Web App
using JavaScript. After you define a rectangle that serves as your
drawing canvas, you can draw straight and curved lines, simple
and complex shapes, graphs, and referenced graphic images. You can add
text, color, shadows, gradients, and patterns. The canvas API also enables
you to save or export the canvas as a .png or .jpeg image file.

To define the drawing area, set the \c{width} and \c{height} of a
\c{<canvas>} element. For example, the following sets a drawing area
with a height of 100 pixels and width of 200 pixels:

\code
<canvas id="mycanvas" width="100" height="200"></canvas>
\endcode

By default, \c{canvas} elements are sized 150 pixels high and 300
pixels wide.  You can also set the size of the canvas using CSS:

\code
canvas { height : 200px; width : 100px; }
\endcode

The \c{canvas} element is transparent and has no visible borders until
you \l{Accessing the Rendering Context}{access the 2D rendering
context}.

Resetting the width or height of an existing canvas erases its
contents and resets all the context properties of the canvas to their
default values.

\section1 Accessing the Rendering Context

The rendering \bold{context} defines the methods and attributes needed
to draw on the canvas. QtWebKit currently supports the two-dimensional
rendering context. The following assigns the canvas rendering context
to a \c{context} variable:

\code
var context = canvas.getContext("2d")
\endcode

The 2d context renders the canvas as a coordinate system whose origin
(0,0) is at the top left corner, as shown in the figure below.
Coordinates increase along the \c{x} axis from left to right and along
the \c{y} axis from top to bottom of the canvas.

\image webkit-guide/canvas_context.gif

\section1 Drawing Shapes

The 2D rendering context supports rectangles, lines, and arcs, which
you can combine to build complex shapes and graphic images.
    \section2 Drawing Rectangles

    The rectangle is the only geometric shape that is built in to the
    canvas API. You can draw an outline of a rectangle, a filled
    rectangle, and a filled rectangle with clear parts. You do not have to
    create a path to draw a rectangle.

    To draw an outline of a rectangle, use the \c{strokeRect()} method.

    To draw a filled rectangle, use the \c{fillRect()} method. The default
    fill color is black.

    To clear part of a filled rectangle, use the \c{clearRect()} method.

    Each method accepts the following series of arguments:

    \list

    \o \c{x} is the position on the canvas to the right of the origin
    (0,0) of the top left corner of the rectangle

    \o \c{y} is the position on the canvas below the origin of the top
    left corner of the rectangle

    \o \c{width} is the width of the rectangle to be drawn

    \o \c{height} is the height of the rectangle to be drawn

    \endlist

    For example, the following code draws concentric rectangles:

    \code
    var context = canvas.getContext("2d");
    canvas.strokeRect(50,50,50,50);
    canvas.fillRect(60,60,30,30);
    canvas.clearRect(70,70,10,10);
    \endcode

    \image webkit-guide/canvas_rectangles.gif

    \section2 Drawing Lines

    To draw a line, you first have to "put your pencil down" on the canvas
    by creating a path. The \c{context.beginPath()} method sets a new path
    in the canvas. Each line that you draw is stored as a sub-path.
    Sub-paths can be closed to form a shape, or they can be left open.
    Each time you want to draw a new shape, you have to call the
    \c{beginPath()} method to reset the current path.

    After calling \c{beginPath()}, you set your starting position on the
    canvas by calling the \c{context.moveTo(x,y)} method. The
    \c{moveTo(x,y)} method creates a new subpath on the canvas that begins
    at the point (x,y).

    To draw a straight line, call the \c{context.lineTo(x,y)} method. This
    adds the point (x,y) to the current subpath and connects it to the
    previous subpath by a straight line. In other words, (x,y) are the
    coordinates of the line\'s endpoint. For example:

    \code
    context.beginPath();
    context.moveTo(10,10);
    context.lineTo(30,30);
    \endcode

    To get the "pencil" to actually draw on the canvas, first use the
    \c{strokeStyle} property to set the color to a value such as black
    (#000):

    \code
    context.strokeStyle(#000);
    \endcode

    (The \c{strokeStyle} property can be a CSS color, a pattern, or a
    gradient.)  Then use the \c{context.stroke()} method to actually draw
    the line on the canvas:

    \code
    context.stroke();
    \endcode

    This produces the image below. The numeric coordinates are added for
    clarity but are not part of the image drawn by the code:

    \image webkit-guide/canvas_lineStrokeTo.gif

    To create a shape, call the \c{context.closePath()} method:

    \code
    context.closePath();
    context.moveTo(10,10);        // starting point
    context.lineTo(30,30);        // specify first line
    context.moveTo(30,30);        // move to end of first line
    context.lineTo(60,10);        // specify second line
    context.moveTo(60,10);        // move to end of second line
    context.lineTo(10,10);        // specify third line to close triangle
    context.strokeStyle("#000");    // use black color for lines
    context.stroke();        // draws the triangle lines on the canvas
    \endcode

    To fill the shape, use the \c{fillstyle} property and the \c{fill()}
    method:

    \code
    context.fillStyle("#FF0000"); // use red color for fill
    context.fill(); // fill the triangle
    \endcode

    The above commands, if coded fully, would create the shape below:

    \image webkit-guide/canvas_closepath.gif

    \bold{NOTE:} It is not necessary to close the path when calling the
    \c{fill()} method. Calling \c{fill()} closes the path and creates the
    completed shape.

    You can draw lines of various widths, endcap types, and joining
    options by configuring the following attributes:

    \list

    \o \c{lineWidth} sets the thickness of the current line. The value can
    be any number greater than 0. For example, \c{context.lineWidth = 10}
    sets the line thickness to 10 units. The default value is 1 unit,
    which is not the same as 1 pixel. Instead, the line is centered on the
    current path, with half its thickness on each side of the path.

    \o \c{lineCap} sets the type of endpoint of the current line. The
    value can be either \c{butt}, \c{square}, or \c{round}. (The
    default value is \c{butt}.)

    \list

    \o \c{butt}- the ends of the line abutt the line guide.

    \o \c{square} adds a box at both ends of the line.

    \o \c{round} adds a semicircle at both ends of the line.

    \endlist

    \o \c{lineJoin} sets the style with which lines are joined. The value
    can be either \c{bevel}, \c{round}, or \c{miter}. (The default value
    is \c{miter}.)

    \list

    \o \c{bevel} flattens the corners at which the lines join

    \o \c{round} rounds the corners at which the lines join

    \o \c{miter}  joins the lines at a single point

    \endlist

    \o \c{miterLimit} sets the miter limit ratio. The value can be any
    number greater than 0. The miter limit ratio determines how far the
    connection point of the outside of the lines can be from the
    connection point of the inside of the lines. (The default value is
    10.)

    \endlist

    \image webkit-guide/canvas_linecap.png

    \section2 Drawing Arcs

    To draw an arc, you begin with the same steps your followed to create
    a line:

    \list 1

    \o Call the \c{context.beginPath()} method to "put your pencil down" on the
    canvas and set a new path.

    \o Call the \c{context.moveTo(x,y)} method to set your starting
    position on the canvas at the point (x,y).

    \o To draw an arc or circle, call the
    \c{context.arcTo(x1,y1,x2,y2,radius)} method. This adds an arc with
    starting point \c{(x1,y1)}, ending point \c{(x2,y2)}, and radius \c{radius} to the
    current subpath and connects it to the previous subpath by a straight
    line.

    \image webkit-guide/canvas_arcTo.png

    \o An alternative way to draw an arc or circle is to call the
    \c{context.arc(x,y,radius,startAngle,endAngle,anticlockwise)}
    method. This adds an arc to the current subpath that lies on the
    circumference of the circle whose center is at the point (x,y) and
    whose radius is \c{radius}.

    \image webkit-guide/canvas_arcTo2.png

    Both \c{startAngle} and \c{endAngle} are measured from the x axis in
    units of radians.

    A complete circle is 360 degrees, or 2\pi radians. A semicircle is 180
    degrees, or \pi radians. The number of radians is the number of
    degrees multiplied by \pi/180, expressed in JavaScript as:

    \code
    var radians = (Math.PI/180)*degrees;
    \endcode

    \image webkit-guide/canvas_startAngle.png

    \c{anticlockwise} has the value \c{TRUE} for each arc in the figure
    above because they are all drawn in the counterclockwise direction.

    \o To create a shape, call the \c{context.closePath()} method. This
    marks the current subpath as closed and draws a straight line from the
    current point to the first point in the path.

    \o To draw only the outline of the shape, call the \c{context.stroke()}
    method.

    \o To fill in the shape, call the \c{context.fill()} method.

    \o To set the color of the fill, set the \c{strokeStyle}. For example,
    the code

    \code
    context.strokeStyle = "#FF0000";
    \endcode

    sets the fill color to red.

    \endlist

    \bold{NOTE:} It is not necessary to close the path if you are going to call
    the \c{fill()} method. The fill closes the path and creates the completed
    shape.

    To create complex shapes, combine lines and arcs:

    \list 1

    \o Call the \c{context.beginPath()} method to "put your pencil down"
    on the canvas and set a new path.

    \o Call the \c{context.moveTo(x,y)} method to set your starting
    position on the canvas at the point (x,y).

    \o Draw any combination of lines and arcs by calling the \c{lineTo},
    \c{arcTo}, \c{arc}, \c{moveTo}, \c{closePath}, \c{stroke}, and
    \c{fill} methods and setting the line attributes and fill colors as
    described above.

    \endlist

    You can also create complex shapes by removing portions of the shapes
    you draw. The \c{clip()} method creates a clipping path that defines
    the area along which your "scissor" will cut. Any parts of the shape
    outside the clipping path are not displayed. To create a complex shape
    using the \c{clip()} method:

    \list 1

    \o Call the \c{context.beginPath()} method to set the clipping path.

    \o Define the clipping path by calling any combination of the
    \c{lineTo}, \c{arcTo}, \c{arc}, \c{moveTo}, and \c{closePath} methods.

    \o Call the \c{context.clip()} method.

    \endlist

    The new shape displays.  The following shows how a clipping path can
    modify how an image displays:

    \image webkit-guide/canvas_clip-complex.png

\section1 Compositing

You can build complex shapes by drawing shapes on top of each
other. It is also possible to draw shapes behind existing shapes and
to mask parts of shapes by using compositing operations. The
\c{globalCompositeOperation} attribute sets the way shapes can be
combined.

The first shape drawn on the canvas to which additional shapes are
added is called the destination shape. The shape drawn on the canvas
afterwards to create the composite image is called the source shape.
The value of the \c{globalCompositeOperation} attribute must be set to
one of the following:

\list

\o \c{source-over} displays the source (newer) shape over the
destination (older) shape unless the source shape is transparent.
(This is the default value)

\o \c{source-in} displays only the portion of the source shape that is
opaque and overlaps the destination shape. Everything else is
transparent.

\o \c{source-out} displays only the portion of the source shape that
does not overlap the destination shape.

\o \c{source-atop} displays only the portion of the opaque source
shape that overlaps the destination shape and the portion of the
destination shape that is not covered by the opaque source shape.

\o \c{destination-over} displays the destination shape over the source
shape. In areas where both shapes are opaque and overlap, the older
shape displays.

\o \c{destination-in} displays only the portion of the destination
shape that is opaque and overlaps the source shape. Everything else is
transparent. The source (newer) shape is not visible.

\o \c{destination-out} displays only the portion of the destination
shape that does not overlap the source shape. The source shape is not
visible.

\o \c{destination-atop} displays only the portion of the opaque
destination shape that overlaps the source shape and the portion of
the source shape that is not covered by the opaque destination shape.

\o \c{lighter} displays both the source and destination shapes. Where
the shapes overlap, the their color values are added, producing a
lighter color.

\o \c{copy} displays only the source shape. The destination shape is
ignored.

\o \c{xor} displays both the source and the destination shapes except
the areas of overlap, in which both shapes are completely transparent.

\endlist

The following figure shows the various compositing effects:

\image webkit-guide/canvas_composite.png

\section1 Saving and Exporting Canvas Drawings as Image Files

You can save or export your canvas drawings as .png or .jpeg image
files by calling the \c{toDataURL()} method:

\code
canvas.toDataURL([type, ...])
\endcode

where:

\list

\o \c{type} is the MIME type to which you want to save or export your
canvas. Possible values are:

\list

\o \c{"image\png"} (Default value)
\o \c{"image\jpeg"}

\endlist

\o\c{...} represents additional arguments that depend on the MIME type.

\list

\o If \c{type} is \c{png}, this argument is \c{" "}
\o If \c{type} is \c{jpeg}, this argument is the desired quality level of the image. The value is a number in the range 0.0 to 1.0, inclusive.

\endlist
\endlist

\section1 Drawing Text

You can draw text on your canvas by setting the following font
attributes on the 2d drawing context:

\list

\o \c{font} refers to any font, expressed the same way as in CSS
properties. This attribute\'s value can include any font style,
variant, weight, size, height, and family. For example:

\code
context.font = "12pt Arial";
\endcode

The default value is \c{10px sans-serif}.

If you set the \c{font} attribute to a relative font size, the browser
multiplies it by the computed font size of the \c{<canvas>} element
itself. For example:

\code
context.font = "200%";
\endcode

\o \c{textAlign} specifies the alignment of the text. The values can
be one of the following:

\list

\o \c{left} for left-aligned text

\o \c{right} for right-aligned text

\o \c{center} for text that is centered within each line

\o \c{start} (default) - the text is aligned at the beginning of the
line. Text is left- or right-justified based on locale-specific
writing method: left when text is left-to-right, right when text is
right-to-left.

\o \c{end} - the text is aligned at the end of the line, either left or
right depending on locale-specific writing method.

\endlist

\o \c{textBaseline} specifies the position at which text is drawn
relative to a baseline. The figure below, from
\l{http://dev.w3.org/html5/canvas-api/canvas-2d-api.html}{the World
Wide Web Consortium}, illustrates the possible values for the
\c{textBaseline} attribute:

\list

\o \c{top} is the top of the em square, which approximates the top of the
glyphs in a font

\o \c{hanging} specifies a hanging baseline, where the tops of some
glyphs are anchored.

\o \c{middle} is the mid-point of the em square

\o \c{alphabetic} (default) is the anchor point of many alphabetic
characters

\o \c{ideographic} is the anchor point of many ideograms, such as the
characters used in the writing systems of many Asian languages

\o \c{bottom} is the bottom of the em square

\endlist

\endlist

\image webkit-guide/canvas_text.png

To draw text on a canvas:

\list 1

\o Set the \c{font} attribute on the drawing context. For example:

\code
context.font = "bold 11px arial"
\endcode

\o Measure the text that you want to draw by calling the \c{measureText}
method:

\code
TextMetrics measureText("Text to draw");
\endcode

where \c{TextMetrics} is the object returned. Its \c{width} attribute
is the width, in pixels, that the "Text to draw" would be when drawn
with the font specified by the \c{font} attribute.

\o Call either of the following methods:

\list

\o \c{fillText} draws the text with the font style specified by the
\c{font} attribute, the alignment specified by the \c{textAlign} attribute,
and the baseline specified by the \c{textBaseline} attribute. For example:

\code
context.fillText("Text to draw",x,y,maximumWidth);
\endcode

where \c{x} and \c{y} are the coordinates at which the drawing begins
(the anchor point), and \c{maximumWidth} is the maximum width of the
text string (optional). If the \c{width} returned in step 2 is larger
than the \c{maximumWidth}, the font is scaled down until the width of
the text string is less than the \c{maximumWidth} specified.

If you don\'t specify the \c{font} attribute, the text inherits the
font size and style of the \c{<canvas>} element itself.

\o \c{strokeText} is the same as the \c{fillText} method, except that
a stroke style is applied to the text instead of a fill style,
creating outlines of glyphs. For example:

\code
context.fillText("Text to stroke",x,y,maximumWidth);
\endcode

\endlist

\endlist

\section1 Working with Images

You can insert existing images onto your canvas, you can scale or crop
them, and you can combine them to create composite images. You can
also draw new images by creating an \c{Image()} object with JavaScript.

To insert an existing image onto a canvas, call the \c{drawImage} method:

\code
context.drawImage(image, dx, dy, dw, dh)
\endcode

where:

\list

\o \c{image} is a reference to an HTML \c{<image>} or \c{<canvas>}
element. The image must be fully loaded before you can draw it on the
canvas. The reference cannot be a URL. Instead, it should be
referenced using standard DOM methods such as \c{document.images()} or
\c{document.getElementById()}.  For example:

\code
<canvas id="demo1" width="100" height="150"></canvas>

var canvas = document.getElementById("demo1");
var context = canvas.getContext("2d");
\endcode


\o \c{dx} is the x coordinate of the upper left corner of the image to be
drawn on the canvas (the destination image)

\o \c{dy} is the y coordinate of the upper left corner of the destination
image

\o \c{dw} is the width of the destination image (optional)

\o \c{dh} is the height of the destination image (optional)

\endlist

If \c{dw} and \c{dh} are not specified, the image retains its source
dimensions when drawn on the canvas. When \c{dw} and \c{dh} are
specified, the image is scaled to width \c{dw} and height \c{dh} when
drawn on the canvas.

If you want to crop the source image, the \c{drawImage} method can be
overloaded with the following arguments:

\code
context.drawImage(image, sx, sy, sw, sh, dx, dy, dw, dh)
\endcode

where:

\list

\o \c{sx} is the x coordinate of the upper left corner of the cropped
source image

\o \c{sy} is the y coordinate of the upper left corner of the cropped
source image

\o \c{sw} is the width of the cropped source image

\o \c{sh} is the height of the cropped source image

\endlist

Use this method if you want to crop the source image to the rectangle
(sx, sy, sw, sh) before drawing it on the canvas. The destination
image will have width dw, height dh, and upper left corner at
coordinates (dx,dy) on the canvas.

To create a new image using JavaScript, create an \c{Image} object and
define its source. Use an \c{onload} event handler to ensure that the
\c{drawImage} method is not called until the image has finished loading.
For example:

\code
var graphic = new Image();
graphic.src = "clipart/graphic.png";
\endcode

The image begins to load.

\code
graphic.onload = function(){
    context.drawImage(graphic,x,y);
};
\endcode

    \section2 Creating Patterns with Images

    You can create patterns with an image by repeating it horizontally,
    vertically, or both. The top left corner of the first image must be
    anchored at the origin of the coordinate space. To repeat an image,
    call the \c{createPattern} method:

    \code
    context.createPattern(image, repetition);
    \endcode

    where:

    \list

    \o \c{image} is a reference to an HTML \c{<image>} or \c{<canvas>}element
    that is repeated to form a pattern. The image must
    be fully loaded before you can draw it on the canvas. The reference
    cannot be a URL. Instead, it should be referenced via standard DOM
    methods such as \o \c{document.images} and \o
    \c{document.getElementById}.  For example:

    \code
    <canvas id="demo1" width="100" height="150"></canvas>

    var canvas = document.getElementById("demo1");
    var context = canvas.getContext("2d");
    \endcode

    \o \c{repetition} is the direction in which the image repeats to form
    the pattern. Possible values are:

    \list

    \o \c{repeat} (default) the image repeats both horizontally and
    vertically

    \o \c{repeat-x} the image repeats horizontally

    \o \c{repeat-y} the image repeats vertically

    \endlist

    \endlist

    The repeated images are the same size as the source image. The
    \c{createPattern} method does not scale the images.

    For example, to create a horizontal pattern of roses, create an
    \c{Image} object to use as a pattern and define its source. Use an
    \c{onload} event handler to ensure that the \c{createPattern} method
    is not called until the image has finished loading. For example:

    \code
    var roses = new Image();
    roses.src = "clipart/roses.jpg";
    \endcode

    The image begins to load.

    \code
    roses.onload = function(){
        var pattern = context.createPattern(roses,repeat-x);
    };
    \endcode

    \image webkit-guide/canvas_pattern.png

\section1 Applying Colors

To draw the outline of a shape in color,
set the \c{strokeStyle} attribute to any valid
\l{http://www.w3schools.com/css/css_colors.asp}{CSS color value}.
The color value can be in hexadecimal notation or in RGB/HSL notation,
as described in \l{Specifying Color and Opacity}.
For example,
either of the following sets a shape\'s outline to red:

\code
context.strokeStyle = "#FF0000"
context.strokeStyle = "rgb(255,0,0)"
\endcode

To fill a shape with color,
set the \c{fillStyle} attribute to a
\l{http://www.w3schools.com/css/css_colors.asp}{CSS color value}.
The color value can be in hexadecimal notation or in RGB/HSL
notation.
For example, either of the following colors a shape\'s interior as
blue:

\code
context.fillStyle = "#0000FF"
context.fillStyle = "rgb(0,0,255)"
\endcode

The
\l{http://www.w3.org/TR/2003/CR-css3-color-20030514/#numerical}{CSS3
Color Module specification} extends both RGB and HSL color models to
include a color\'s opacity,
referred to as its \bold{alpha}.
These extended models are known as RGBA and HSLA.
There are no hexadecimal notations for RGBA and HSLA values.
The following specifies varying levels of opacity for a blue shape:

\code
context.fillStyle = rgba(0, 0, 255, 0)        // transparent
context.fillStyle = rgba(0, 0, 255, 0.5)    // semi-transparent
context.fillStyle = rgba(0, 0, 255, 1)        // opaque
\endcode

When you set the \c{context.strokeStyle} or \c{context.fillStyle}
attributes,
whatever value you set becomes the default value for all subsequently
drawn shapes,
until you set a new value.

    \section2 Applying Gradients

    A gradient is a smooth transition between colors. There are two types
    of gradients: linear and radial.

    A linear gradient transitions the color along a line between two
    points. To create a linear gradient, call the \c{createLinearGradient}
    method:

    \code
    createLinearGradient(x0, y0, x1, y1)
    \endcode

    where \c{(x0, y0)} is the starting point and \c{(x1, y1)} is the ending point
    for the linear gradient.

    A radial gradient transitions the color along a cone between two
    circles. To create a radial gradient, call the \c{createRadialGradient}
    method:

    \code
    createRadialGradient(x0, y0, r0, x1, y1, r1)
    \endcode

    where:

    \list

    \o \c{(x0, y0, r0)} represents the starting circle, whose origin is \c{(x0,
    y0)} and whose radius is \c{r0}.

    \o \c{(x1, y1, r1)} represents the ending circle, whose origin is \c{(x1, y1)}
    and whose radius is \c{r1}.

    \endlist

    Gradients must have two or more \bold{color stops}, representing color
    shifts positioned from 0 to 1 between to the gradient\'s starting and
    end points or circles:

    \code
    addColorStop(position,color)
    \endcode

    where:

    \list

    \o \c{position} specifies the position of the color within the already
    defined starting and end points or circles, expressed as a number
    from 0 to 1.

    \o \c{color} specifies the CSS color at that position.

    \endlist

    For example, to define a gradient that varies from red to blue
    horizontally along a rectangular area:

    \list 1

    \o Create a gradient object:

    \code
    var redbluegradient = context.createLinearGradient(0,0,100,0);
    \endcode

    \o Define the color stops:

    \code
    redbluegradient.addColorStop(0, "rgb(255,0,0)"); // red at the left side of the rectangle
    redbluegradient.addColorStop(1, "rgb(0,0,255)"); // blue at the right side of the rectangle
    \endcode

    \o Draw the shape and set a \c{fillStyle} or \c{strokeStyle}:

    \code
    context.fillStyle = redbluegradient;
    context.fillRect(0,0,100,150);
    \endcode

    \endlist

    To define a gradient that varies from red to blue vertically along a
    rectangle:

    \list 1

    \o Create a gradient object:

    \code
    var redbluegradient = context.createLinearGradient(0,0,0,150);
    \endcode

    \o Define the color stops:

    \code
    redbluegradient.addColorStop(0, "rgb(255,0,0)"); // red at the top of the rectangle
    redbluegradient.addColorStop(1, "rgb(0,0,255)"); // blue at the bottom of the rectangle
    \endcode

    \o Draw the shape and set a \c{fillStyle} or \c{strokeStyle}:

    \code
    context.fillStyle = redbluegradient;
    context.fillRect(0,0,100,150);
    \endcode

    \endlist

    \bold{NOTE:} A canvas gradient\'s color stops behave slightly
    differently than those used within non-canvas \l{Gradients}{CSS3
    Webkit gradients}.  Webkit gradients specify mandatory \c{from} and
    \c{to} colors, with optional \c{color-stop} values for additional
    color shifts within the overall range of the gradient. For canvas
    gradients, even the initial and final colors are defined as color
    stops.

    \section2 Applying Shadows

    To add a shadow effect to a drawing on a canvas, set the following
    attributes:

    \list

    \o \c{shadowColor} sets the color of the shadow. The value can be any
    CSS color value. The default value is transparent black (\c{"rgba(0,0,0,0)"}).

    \o \c{shadowBlur} sets the amount of blur in the shadow, in
    pixels. The value can be any positive number or 0. A value of 0
    produces a sharp shadow with no blur.

    \o \c{shadowOffsetX} sets the number of pixels the shadow extends
    horizontally from the object drawn. If this value is a positive
    number, the shadow extends to the right of the object. If negative,
    the shadow extends to the left of the object. The default value is 0
    pixels.

    \o \c{shadowOffsetY} sets the number of pixels the shadow extends
    vertically from the object drawn. If this value is a positive number,
    the shadow extends below the object. If negative, the shadow extends
    above the object. The default value is 0 pixels.

    \endlist

    The following example code adds a semi-transparent black shadow to the
    bottom right of a blue rectangle:

    \code
    var context        = canvas.getContext("2d");
    context.shadowOffsetX    = 5;
    context.shadowOffsetY    = 5;
    context.shadowBlur    = 10;
    context.shadowColor    = "rgba(0,0,0,0.5)";
    context.fillStyle    = "#0000FF";
    context.fillRect    = (0,0,100,50)
    \endcode

\section1 Transforming Graphics

When drawing shapes and paths, you can translate the canvas\'s origin,
rotate the canvas around the origin, scale the units in the canvas
grid, and modify the transformation matrix directly.

    \section2 Translating the Canvas Origin

    Translating the origin enables you to draw patterns of different
    objects on the canvas without having to measure the coordinates
    manually for each shape. To translate the origin of the canvas, use
    the \c{translate} method:

    \code
    context.translate(x,y);
    \endcode

    where:

    \list

    \o \c{x} is the horizontal distance that the origin is translated, in
    coordinate space units

    \o \c{y} is the vertical distance that the origin is translated, in
    coordinate space units

    \endlist

    \section2 Rotating the Canvas

    To rotate the canvas around the current origin, call the \c{rotate()}
    method:

    \code
    context.rotate(angle);
    \endcode

    where \c{angle} is the clockwise rotation angle in radians.

    The number of radians is the number of degrees multiplied by \pi/180,
    expressed in JavaScript as:

    \code
    var radians = (Math.PI/180)*degrees;
    \endcode

    \image webkit-guide/canvas_rotate.png

    \section2 Scaling the Canvas Grid

    To increase or decrease the size of each unit in the canvas grid, call
    the \c{scale} method:

    \code
    context.scale(x,y);
    \endcode

    where:

    \list

    \o \c{x} is the scale factor in the horizontal direction

    \o \c{y} is the scale factor in the vertical direction

    \endlist

    The scale factors are in multiples. For example, \c{scale(2.0, 0.5)} would
    double the horizontal size of an object drawn on the canvas and half
    its vertical size, as shown below:

    \image webkit-guide/canvas_scale.png

    \section2 Manipulating the Transformation Matrix

    Modifying the transformation matrix directly enables you to perform
    scaling, rotating, and translating transformations in a single step.

    The transformation matrix is an affine transformation matrix from
    linear algebra. Affine transformations preserve colinearity and
    relative distance in the transformed coordinate space. This means that
    points in a line remain in a line, parallel lines remain parallel, and
    the distance between lines and objects maintains the same ratio, even
    if a scale factor is applied. Repositioning by translation, rotation,
    or skewing is also possible.

    Each point on the canvas is multiplied by the matrix before anything
    is drawn. The HTML5 canvas API defines the transformation matrix as:

    \image webkit-guide/canvas_math.png

    where:

    \list

    \o \c{a} is the scale factor in the horizontal (x) direction

    \image webkit-guide/canvas_scalex.png

    \o \c{c} is the skew factor in the x direction

    \image webkit-guide/canvas_skewx.png

    \o \c{e} is the translation in the x direction

    \image webkit-guide/canvas_translate.png

    \o \c{b} is the skew factor in the y (vertical) direction

    \image webkit-guide/canvas_skewy.png

    \o \c{d} is the scale factor in the y direction

    \image webkit-guide/canvas_scaley.png

    \o \c{f} is the translation in the y direction

    \image webkit-guide/canvas_translatey.png

    \o the last row remains constant

    \endlist

    The scale factors and skew factors are multiples; \c{e} and \c{f} are
    coordinate space units, just like the units in the \c{translate(x,y)}
    method.

    The rotation transformation matrix is as follows:

    \image webkit-guide/canvas_math_rotate.png

    where the angle of rotation is in radians.

    \sa
    \l{http://www.senocular.com/flash/tutorials/transformmatrix/}{senocular.com}
    for a good explanation of how transformation matrices are used
    identically within Flash.

\section1 Canvas Animations

You can animate a canvas drawing by repeatedly redrawing the canvas
for each frame and translating,
rotating,
skewing,
and scaling the drawn objects.

To draw each frame by employing the HTML5 canvas API,
you should define the original canvas state and save it for future
reference.
The drawing context maintains a stack of drawing states.
Each state consists of the current transformation matrix,
current clipping region,
and current values of the following attributes:
\c{strokeStyle},
\c{fillStyle},
\c{globalAlpha},
\c{lineWidth},
\c{lineCap},
\c{lineJoin},
\c{miterLimit},
\c{shadowOffsetX},
\c{shadowOffsetY},
\c{shadowBlur},
\c{shadowColor},
\c{globalCompositeOperation},
\c{font},
\c{textAlign},
and
\c{textBaseline}.
The current path and the current bitmap are NOT part of the drawing
state.
The path can be reset only by invoking the \c{beginPath()} method.
The current bitmap is a property of the canvas,
not of the context.

To save the original canvas state,
call the \c{save()} method:

\code
context.save();
\endcode

Before drawing each new frame,
you must clear the canvas:

\code
canvas.clearRect(x,y,width,height);
\endcode

where:

\list

\o \c{x} is the position of the top left corner of the canvas on the
horizontal axis

\o \c{y} is the position of the top left corner of the canvas on the
vertical axis

\o \c{width} is the width of the canvas

\o \c{height} is the height of the canvas

\endlist

Draw the new frame using any of the methods provided by the canvas
API.
Then save it by calling the \c{save()} method.

If you wish to return to the state of the original frame as the basis
for each new frame that you draw,
call the \c{context.restore()} method.

To execute the drawing methods repeatedly,
use the standard JavaScript-based animation technique,
calling the \c{setInterval()} and \c{clearInterval()} methods.
The following shows how to execute an animation function every 50
milliseconds (corresponding to 20 times per second,
a typical animation frame rate),
then subsequently halt the animation:

\code
var id = setInterval(functionName, 50);
clearInterval(id);
\endcode

\bold{See Also:}

\list

\o
\l{http://www.canvasdemos.com/2009/10/09/html-5-canvas-animation/}{CanvasDemos.com:
animated cartoon},
which discusses how to use Canvas as an animation framework.

\o
\l{http://blog.nihilogic.dk/2009/02/html5-canvas-cheat-sheet.html}{nihilogic.dk:
HTML5 Canvas Cheat Sheet}

\endlist

*/