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 module of the Qt Toolkit.
**
** $QT_BEGIN_LICENSE:BSD$
** 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 QtWebKit Guide - Canvas Graphics

\chapter Canvas Graphics
This section of the \l{QtWebKit Guide} serves as an introduction to the
\l{HTML5 Canvas API} features of QtWebKit.

The \l{HTML5 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 \e{"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 Cartesian point \c{(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 \e{pencil} to actually draw on the canvas, first use the
    \c{strokeStyle} property to set the color to a value such as black
    (\c{#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 commands, if coded fully, would create the shape below:

    \image webkit-guide/canvas_closepath.gif

    \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 \c 0. For example, \c{context.lineWidth = 10} sets
    the line thickness to \c 10 units. The default value is \c 1 unit, which is
    not the same as \c 1 \e 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 \e{miter limit ratio}. The value can be any
    number greater than \c 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 \c 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 \e{"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 \c 360 degrees, or 2\pi radians. A semicircle is \c 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

    \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 \e{"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 \e{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 \e{destination} shape. The shape drawn on the canvas afterwards to
create the composite image is called the \e{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{HTML5 Canvas API}, 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 \c{"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
\c dw, height \c dh, and upper left corner at coordinates \c{(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
        \list
        \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
        \endlist
    \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{CSS Color Value}{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{CSS Color
Value}{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{CSS3 Color Module specification} extends both RGB and HSL color models to
include a color's opacity, referred to as its \e{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 \e{color stops}, representing color
    shifts positioned from \c 0 to \c 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 \c 0
    to \c 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

    \note A canvas gradient's color stops behave slightly differently than those
    used within non-canvas \l{Gradients}{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 \e{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 \l{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 \c angle of rotation is in radians.

    \bold{See Also:}
    \l{http://www.senocular.com/flash/tutorials/transformmatrix/}{senocular.com}
    for a good explanation of how transformation matrices are used
    identically within Adobe 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:
\list
\o\c{strokeStyle}
\o\c{fillStyle}
\o\c{globalAlpha}
\o\c{lineWidth}
\o\c{lineCap}
\o\c{lineJoin}
\o\c{miterLimit}
\o\c{shadowOffsetX}
\o\c{shadowOffsetY}
\o\c{shadowBlur}
\o\c{shadowColor}
\o\c{globalCompositeOperation}
\o\c{font}
\o\c{textAlign}
\o\c{textBaseline}
\endlist
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 \c 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}

\o \l{QtWebKit Guide} -back to the main page
\endlist

*/