|CGI Programming with Perl|
Outputting Image Data
Generating PNGs with GD
Additional GD Modules
Throughout this book we have seen many examples of CGI scripts generating dynamic output. However, in almost all cases, the output has been HTML. Certainly this is the most common format your scripts will generate. However, CGI scripts can actually generate any type of format, and in this chapter we will look at how we can dynamically generate images.
Generating images dynamically has many uses. One of the most common is to generate graphs. If you have a data source that is continually changing, such as the results of an online survey, a CGI script can generate a graph that presents a visual snapshot of this data.
There are also times when generating images dynamically makes less sense. It is much less efficient to generate an image dynamically than for your web server to serve the image from an image file. Thus, just because some of these tools allow you to generate really cool graphics dynamically doesn't mean you must use them only in a dynamic context. Unless the images you generate are based upon data that changes, save the image to a static file and serve that instead.
This chapter presents a broad overview of the different tools available for generating dynamic images online, and includes references with each for finding more information. The goal of this chapter is to explain techniques for generating images dynamically and familiarize you with the most popular tools available to you. A full description of many of these tools along with others is available in a book of its own, Programming Web Graphics with Perl and GNU Software by Shawn Wallace (O'Reilly & Associates, Inc.).
Let's first review the image formats that are used online today. The most common image formats, of course, are GIF and JPEG, which every graphical web browser supports. Other file formats that we will discuss in this chapter include PNG and PDF.
The Graphics Interchange Format (GIF ) was created by CompuServe and released as an open standard in 1987. It quickly became a very popular image format and, along with JPEG, became a standard format for images on the Web. GIF files are typically quite small, especially for images with few colors, which makes them well suited for transferring online.
GIF only supports up to 256 colors, but it works well for text and images, such as icons, which do not have many colors but have sharp details. The compression algorithm that GIF uses, LZW, is lossless, which means that no image quality is lost during compression or decompression, allowing GIF files to accurately capture details.
The GIF file format has been extended to support basic animation, which can loop. The moving banner ads that you see online are typically animated GIF files. GIF files can also have a transparent background by specifying a single color in the image that should be displayed as transparent.
Unfortunately, CompuServe and others apparently failed to notice that LZW, the compression algorithm used by GIF, was actually patented by Unisys in 1983. Unisys reportedly discovered that GIF uses LZW in the early 1990s and in 1994 CompuServe and Unisys reached a settlement and announced that developers who write software supporting GIF must pay a licensing fee to Unisys. Note that this does not include web authors who use GIF files or users who browse them on the Web.
This turn of events created quite a stir among developers, especially open source developers. As a result, CompuServe and others developed the PNG format as a LZW-free successor to GIF; we'll discuss PNG below. However, GIF remains a very popular file format, and PNG is not supported by all browsers.
As a result of the LZW licensing issue, the tools we discuss in this chapter provide very limited support for GIF files, as we will see.
The Portable Network Graphic (PNG ) format was created as a successor to the GIF format. It adds the following features over GIF:
PNG uses an efficient compression algorithm that is not LZW. In most cases, it achieves slightly better compression than the LZW algorithm.
PNG supports images in any of three modes: images with a limited palette of 256 or fewer colors, 16-bit grayscale images, and 48-bit true color images.
PNG supports alpha channels, which allows varying degrees of transparency.
PNG graphics have a better interlacing algorithm that allows users to make out the contents of the image as it downloads much faster than with a GIF.
For additional differences, as well as an demonstration of the difference between the PNG and GIF interlacing, visit http://www.cdrom.com/pub/png/pngintro.html.
Unfortunately, many browsers do not support PNG images. Of those that do, many do not support all of its features, such as multiple levels of transparency. Support for PNG should continue to increase, however, and older browsers that do not support it will eventually be upgraded.
PNG does not support animations.
The Joint Photographic Experts Group (JPEG ) is a standards body created to generate an image format for encoding continuous tone images. Their JPEG standard actually discusses a very general method for still image compression and not a file format. The file format that people typically think of as a JPEG is actually JFIF, the JPEG File Interchange Format . We will stick with the more familiar term and also refer to a JFIF file as a JPEG file.
JPEG files are ideal for encoding photographs. JPEG supports full, 24-bit color but it uses a lossy compression algorithm, which means that each time the file is compressed, detail is lost. Because the encoding for JPEG files is done in blocks, it is most noticeable in images that have very sharp details, such as text and line art. These details may appear blurred in a JPEG file.
JPEG files have no support for animation or transparency.
Adobe's Portable Document Format (PDF ) is more than just an image format. It is actually a language derived from PostScript that can include text, basic shapes, line art, and images, as well as numerous other elements. Unlike images, which are typically displayed within an HTML file, PDF files are typically standalone documents, and users use a browser plug-in or external application such as Adobe Acrobat to view them.
|12.3. Inverted Index Search||13.2. Outputting Image Data|
Copyright © 2001 O'Reilly & Associates. All rights reserved.