Welcome to Teach Yourself CGI Programming with Perl in a Week! This is going to be a very busy week. You will need all seven days, but at the end of the week you will be ready to create interactive Web sites using your own CGI programs. This book does not assume that you have experience with the programming language Perl and makes very little assumptions about prior programming experience.
This book does assume that you already have been on the Internet and understand what a Web page is. You do not have to be a Web page author to understand this book. A basic understanding of HTML will be helpful, however. This book spends significant time explaining how to use the HTML Form tag and its components to create Web forms for getting information from your Web clients.
As new topics are introduced throughout the book, most will include an example. And with each new programming example will come a detailed analysis of the new CGI features in that example. CGI programming is a mixture of understanding and using the Hyper-Text Mark Up Language (HTML), the Hyper-Text Transport Protocol (HTTP), and writing code. You must follow the HTML and HTTP specifications, but you can use any programming language with which you are comfortable. For most applications, I recommend Perl.
This book is written primarily for the Unix environment. Because Perl works on any platform and the HTTP and HTML specifications can work on any platform, what you learn from this book can apply to non-Unix operation systems.
However, most of the Net right now is Unix based. "Why is that?" you might ask. Well, it has a lot to do with Unix's more than 20 years of dominance in networked environments. Like everything else in the computer industry, I'm sure this will change, but Unix is the platform of choice for Internet applications, at least for now. So this book assumes that you are programming on a Unix server. Your WWW server probably is NCSA, CERN, or some derivative of these two—like Apache. If you are using some other server, like Netscape's secure server or a Windows NT server, don't despair. Most of this book applies to your environment also.
In this chapter, you will learn the basics of how to install your CGI programs, and you will get an overview of how they work with your server. You also will learn how to avoid some of the common mistakes that come up when you are starting out with CGI programming.
In particular, you will learn about the following:
By the way, you should read this book sequentially by chapter number. Each chapter builds on the knowledge of the preceding chapter.
What is CGI programming anyway? What is the BIG DEAL?? And why the heck is it called a gateway?
Very good questions. Ones that bugged me early on and ones that still seem to get asked quite frequently.
CGI programming involves designing and writing programs that receive their starting commands from a Web page—usually, a Web page that uses an HTML form to initiate the CGI program. The HTML form has become the method of choice for sending data across the Net because of the ease of setting up a user-interface using the HTML Form and Input tags. With the HTML form, you can set up input windows, pull-down menus, checkboxes, radio buttons, and more with very little effort. In addition, the data from all these various data-entry methods is formatted automatically and sent for you when you use the HTML form. You will learn about the details of using the HTML form in Chapters 4, "Using Forms to Gather and Send Data," and 5, "Decoding Data Sent to Your CGI Program."
CGI programs don't have to be started by a Web page, however. They can be started as the result of a Server Side Include execution command (covered in detail in Chapter 3, "Using Server Side Include Commands"). You even can start a CGI program from the command line. But a CGI program started from the command line probably will not act the way you expect or designed it to act. Why is that? Well, a CGI program runs under a unique environment. The WWW server that started your CGI program creates some special information for your CGI program and it expects some special responses back from your CGI program.
Before your CGI program is initiated, the WWW server already has created a special processing environment for your CGI program in which to operate. That environment includes translating all the incoming HTTP request headers (covered in Chapter 2, "Understanding How the Server and Browser Communicate") into environment variables (covered in Chapter 6, "Using Environment Variables in Your Programs") that your CGI program can use for all kinds of valuable information. In addition to system information, like the current date, is information about who is calling your CGI program, where your program is being called from, and possibly even state information to help you keep track of a single Web visitor's actions. (State information is anything that keeps track of what your program did the last time it was called.)
Next, the server tries to determine what type of file or program it is calling because the server must act differently based on the type of file it is accessing. So, your WWW server first looks at the file extension to determine whether it needs to parse the file looking for Server Side Include commands, execute the Perl interpreter to compile and interpret a Perl program, or just generate the correct HTTP response headers and return an HTML file.
After your server starts up your Server Side Include or CGI program (or even HTML file), it expects a specific type of response from the Server Side Include or CGI program. If your server is just returning an HTML file, it expects that file to be a text file with HTML tags and text in it. If the server is returning an HTML file, the server is responsible for generating the required HTTP response headers, which tell the calling browser the status of the browser's request for a Web page and what type of data the browser will be receiving, among other things.
The Server Side Include (SSI) file works almost like a regular HTML file. The only difference is that with an SSI file, the server must look at each line in the file for special SSI commands. If it finds an SSI command, it tries to execute it. The output from the executed SSI command is inserted into the returned HTML file, replacing the special HTML syntax for calling an SSI command. The output from the SSI command will appear within the HTML text just as if it were typed at the location of the SSI command. SSI commands can include other files, execute system commands, and perform many useful functions. The server uses the file extension of the requested Web page to determine whether it needs to parse a file for SSI commands. SSI files typically have the extension .shtml.
If the server identifies the file as an executable CGI program, it executes the program as appropriate. After the server executes your CGI program, your CGI program normally responds with the minimum required HTTP response headers and then some HTML tags. If your CGI program is returning HTML, it should output a response header of content-type: text/html. This gives the server enough information to generate any other required HTTP response headers.
After all that explanation, what is CGI programming? CGI programming is writing the programs that receive and translate data sent via the Internet to your WWW server. CGI programming is using that translated data and understanding how to send valid HTTP response headers and HTML tags back to your WWW client.
The big deal in all this is a brand new dynamic programming environment. All kinds of new commerce and applications are going to occur over the Internet. You can't do this with just HTML. HTML by itself makes a nice window, but to do anything more than look pretty requires programming, and that programming must understand the CGI environment.
Finally, just why is it called gateway? Well, quite often, your programs will act as a gateway or interface program between other larger applications. CGI programs often are written in scripting languages like Perl. Scripting languages really are not meant for large applications. So, your program could translate and format the data being sent to it from applications such as on-line catalogs, for example. This translated data then would be passed to some type of database program. The database program would do the necessary operations on its database and return the results to your CGI program. Your CGI program then could reformat the returned data as needed for the Internet and return it to the on-line catalog customer, thus acting as a gateway between the HTML catalog, the HTTP request/response headers, and the database program. I'm sure you can think of other more cool examples, but this one probably will be pretty common in the near future.
Already you can see a lot of interaction between the HTTP request/response headers, HTML, and your CGI programs. Each of these topics is covered in detail in this book, but you should understand how these pieces fit together to create the entire CGI environment.
HTML, HTTP, and your CGI program have to work closely together to make your on-line Internet application work. The HTML code defines the way the user sees your program interface, and it is responsible for collecting user input. This frequently is referred to as the Human Computer Interface code. It is the window through which your program and the user interact. HTTP is the transport mechanism for sending data between your CGI program and the user. This is the behind-the-scenes director that translates and sends information between your Web client and your CGI program. Your CGI program is responsible for understanding both the HTTP directions and the user requests. The CGI program takes the requests from the user and sends back valid and useful responses to the Web client who is clicking away on your HTML Web page.
HTML, the Hyper-Text Mark-Up Language, is designed primarily for formatting text. HTML is basically a typesetting language that tells the computer what color to make the text, where to put text, how large to make the text, and what shape the text should be. It's not much different than most other typesetting languages, except that it doesn't have as many typesetting options as most simple WYSIWYG (What You See Is What You Get) editors, such as Microsoft Word. So how does it get involved with your CGI program? The primary method is through the HTML Form tags. It is not required, however, that your CGI program be called through an HTML form; your CGI program can be invoked through a simple hypertext link using the anchor (<a>) tag—something like this:
<a href="A CGI program"> Some text </a>
The CGI program in this hypertext reference or link would be called (or activated) in a manner similar to being called from an HTML form.
You even can use a link to pass extra data to your CGI program. All you have to do is add more information after the CGI program name. This information usually is referred to as extra path information, but it can be any type of data that might help identify to your CGI program what it needs to do.
The extra path information is provided to your CGI program in a variable call PATH_INFO, and is any data after the CGI program name and before the first question mark (?) in the href string. If you include a question mark (?) after the CGI program name and then include more data after the question mark, the data goes in a variable called the QUERY_STRING. Both PATH_INFO and QUERY_STRING are covered in Chapter 6.
So to put this all into an example, suppose that you create a link to your CGI program that looks like the following:
<a href=www.practical-inet.com/cgibook/chap1/program.cgi/extra-path-info?test=test-number-1> A CGI Program </a>
Then when you select the link A CGI program, the CGI program named program.cgi is activated. The environment variable PATH_INFO is set to extra-path-info and the QUERY_STRING environment variable is set to Test=Test-number-1.
Usually, this is not considered a good way to send data to your CGI program. First it's harder for the programmer to modify data hard coded in an HTML file because it cannot be done on-the-fly. Second, it is easier to modify data for the Web page visitor who is a hacker. Your Web page visitor can download the Web page onto his own computer and then modify the data your program is expecting. Then he can use the modified file to call your CGI program. Neither of these scenarios seems very pleasant. Many other people felt the same way, so this is where the HTML form comes in. Don't completely ignore this method of sending data to your program. There are valid reasons for using the extra-path-info variables. The image map program, for example, uses extra-path-info as an input parameter that describes the location of map files. Image maps are covered in Chapter 9, "Livening Up Your Web Page with Image Maps."
The HTML form is responsible for sending dynamic data to your CGI program. The basics outlined here are still the same. Data gets passed to the server for use by your CGI program, but the way you build your HTML form defines how that data will be sent, and your browser does most of the data formatting for you.
The most important feature of the HTML form, however, is the capability of the data to change based on user input. This is what makes the HTML Form tag so powerful. Your Web page client can send you letters, fill out registration forms, use clickable buttons and pull-down menus to select merchandise, or fill out a survey. With a clear understanding of the HTML Form tag, you can build highly interactive Web pages. Because this topic is so important, it is covered in Chapters 4 and 5, and the hidden field of the HTML form is explained Chapter 7, "Building an On-Line Catalog."
So, to sum up, HTML and, in particular, the HTML Form tag, are responsible for gathering data and sending it to your CGI program.
If HTML is responsible for gathering data to send to your CGI program, how does it get there? The data gathered by the browser gets to your CGI program through the magic of the Hyper-Text Transport Protocol request header (HTTP header). The HTML tags tell the browser what type of HTTP header to use to talk to the server, your CGI program. The basic HTTP headers for beginning communication with your CGI program are Get and Post.
If the HTML tag calling your program is a hypertext link, such as
<a href="www.domain.com/progam.cgi">, call a CGI program </a>
then the default HTTP request method Get is used to communicate with your CGI program. If, instead of using a hypertext link to your program, you use the HTML Form tag, then the Method attribute of the Form tag defines what type of HTTP request header is used to communicate with your CGI program. If the Method field is missing or set to Get, the HTTP method request header type is Get. If the Method attribute is set to Post, then a Post Method request header is used to communicate with your CGI program. (The Get and Post methods are covered in Chapters 4 and 5.)
Once the method of sending the data is determined, the data is formatted and sent using one of two means. If the Get method is used, the data is sent via the Uniform Resource Identifier (URI) field. (URI is covered in Chapter 2.) If the Post method is used, the data is sent as a separate message, after all the other HTTP request headers have been sent.
After the browser determines how it is going to send the data, it creates an HTTP request header identifying where on the server your CGI program is located. The browser sends to the server this HTTP request header. The server receives the HTTP request header and calls your CGI program. Several other request headers can go along with the main request header to give the server and your CGI program useful information about the browser and this connection.
Your CGI program now performs some useful function and then tells the server what type of response it wants to send back to the server.
So where are we so far? The data has been gathered by the browser using the format defined by the HTML tags. The data/URI request has been sent to the server using HTTP request headers. The server used the HTTP request headers to find your CGI program and call it. Now your CGI program has done its thing and is ready to respond to the browser. What happens next? The server and your CGI program collaborate to send HTTP response headers back to the browser.
What about the data—the Web page—your CGI program generated? Well, that is what the HTTP response headers are for. The HTTP response headers describe to the browser what type of data is being returned to the browser.
Your CGI program can generate all the HTTP response headers required for sending data back to the client/browser by calling itself a non-parsed header CGI program. If your CGI program is an NPH-CGI program, the server does not parse or look at the HTTP response headers generated by your CGI program. The HTTP request headers are sent directly to the requesting browser, along with data/HTML generated by your CGI program.
The more common form of returning HTTP response headers, however, is for your CGI program to generate the minimum required HTTP request headers; usually, just a Content-Type HTTP response header is required. The server then parses, or looks for, the response header your CGI program generated and determines what additional HTTP response headers should be returned to the browser.
The Content-Type HTTP response header identifies to the browser the type of data that will be returned to the browser. The browser uses the Content-Type response header to determine the types of viewers to activate so the client can view things like in-line images, movies, and HTML text.
The server adds the additional HTTP response headers it knows are required and then bundles up the set of the headers and data in a nice TCP/IP package and sends it to the browser. The browser receives the HTTP response headers and displays the returned data as described by the HTTP response headers to your customer, the human.
So now you have the whole picture (which you will learn about in detail throughout the book), made up of the HTML used to format the data and the HTTP request and response headers used to communicate between the browser and server what type of data is being sent back and forth. Among all this is your very cool CGI program, aware of what is going on around it and driving the real applications in which your Web client really is interested.
What about your CGI program? What is it and how does it fit into this scenario? Well, your CGI program can be anything you can imagine. That is what makes programming so much fun. Your CGI program must be aware of the HTTP request headers coming in and its responsibility to send HTTP response headers back out. Beyond that, your CGI program can do and work in any manner you choose.
For the purposes of this book, I concentrate on CGI programs that work on Unix platforms, and I use the Perl programming language. I focus on the Unix platform because that is the platform of choice on the Net at this time. The most popular WWW servers are the NCSA httpd, CERN, Apache, and Netscape servers; all these Web servers sit most comfortably on Unix operating systems. So, for the moment, most platforms on which CGI programs are developed are Unix servers. It just makes sense to concentrate on the operating system on which most of the CGI applications are required to run.
But why Perl? Well, wouldn't it be nice to work with a language that you didn't have to compile? No messing with painful linker commands. No compilation steps at all. Just type it in and it's ready to go. What about a language that is free? Easy to get a hold of and available on about any machine on the Net? How about a language that works well with and even looks like C, arguably the most popular programming language in the world? And wouldn't it be nice if that language worked well with the operating system, making each of your system calls easy to implement? And what about a programming language that works on almost any operating system? That way, if you change platforms from Unix to Windows, NT, or Mac, your programs still run. Heck, why not just ask for a language that's easy to learn and for which there is a ton of free technical help? Ask for it. You've got it! Did that sound like an advertisement? And no, I don't have any vested interest in Perl.
Perl is rapidly becoming one of the most popular scripting languages anywhere, because it really does satisfy most of the needs outlined here. It's free, works on almost any platform, and runs as soon as you type it in. As long as you don't have any bugs...
Perl is an excellent choice for all these reasons and more. The more is probably what makes the language so popular. If Perl could do all those wonderful things and turned out to be hard to work with, slow, and insecure, it probably would have lost the popularity war. But Perl is easy to work with, has built-in security features, and is relatively fast.
In fact, Perl was designed originally for working with text, generating reports, and manipulating files. It does all these things fairly well, and fairly easily. Larry Wall and Randal L. Schwartz of Programming Perl state that "The pattern matching and textual manipulation capabilities of Perl often out-perform dedicated C programs."
In addition, Perl has a lovely data structure called the associative array that you can use for database manipulation. The designers of Perl also thought of security when they built the language. It has built-in security features like data-flow tracing, which enables you to find out where insecure data originated. This capability often prevents insecure operations before they can occur.
Most of these features will not be covered in this book. If you have never used Perl or are new to programming, however, this book will take the time to show you how to use Perl to develop CGI programs. After you get the basics from this book, you should be able to understand other Perl CGI programs on the Net. As an added bonus, by learning Perl, you get an introduction to Unix and C for free. These reasons were enough to make me want to learn Perl and are the reasons you will use Perl throughout this book.
At this point, you have a good overview of CGI programming and how the different pieces fit together. As you go through the book, most of the topics in these first two sections will be covered again in more detail and with specific examples. The next steps now are for you to learn more about your server, how to install CGI programs, and what makes CGI programming so different from other programming paradigms.
The first thing you need to learn is how to get around your server. If you have a personal account with an Internet service provider, your personal directory should be based on your user name. In my case, I have a personal account with an Internet service provider and a business account from which I manage multiple business Web pages. Your personal account probably is similar to mine; I can build Web pages for Internet access under a specific directory called public-web. The name isn't really important—just the concept of having a directory where specific operations are allowed.
Usually, you will find that your server is divided into two directory trees. A directory tree consists of a directory and the subdirectories below the main directory. Most Unix Web servers separate their users from the system administrative files by creating separate directory trees called the server root and the document root.
The server root contains all the files for which the Web Master or System Administrator is responsible. You probably will not be able to change these files, but there are several of them you will want to be aware of because they provide valuable information about where your programs can run and what your CGI programs are allowed to do. Below the server root are two subdirectories that you should know about. Those directories, located on the NCSA server, usually are called the log directory and the conf directory. If you are not working on an NCSA server, the CERN and other servers have a similar directory structure with slightly different names.
The log directory is where all the log files are kept. Within the log directory are your error log files. Error log files keep track of each command from your CGI, Server Side Include commands, and HTML files that generate some type of error. When you are having problems getting something to work, the error log file is an excellent place from which to start your debugging. Usually, the file begins with err. On my server, the error log file is called error.log. Another log file you can make good use of is the access.log file. This file contains each file that was accessed by a user. This file often is used to derive access counts for your Web page. Building counters is discussed in Chapter 10, "Keeping Track of Your Web Page Visitors." Also in your log directory is a list of each of the different types of browsers accessing your Web site. On my server, this file is called the referer.log. You can use this information to direct a specific browser to Web pages written just for browsers that can or can't handle special HTML extensions. Redirecting a browser based on the browser type is discussed in Chapter 2. That's just the what's in the log directory. In addition to the log files are the configuration files under the conf directory.
The conf directory contains, in addition to other files, the access.conf and srm.conf files. Understanding these files helps you understand the limitations, or lack there of, placed on your CGI programs. Both these files are covered in more detail in Chapter 12, "Security." This introduction is only intended to familiarize you with their purposes and general layouts.
The access.conf file is used to define per-directory access control for the entire document root. Any changes to this file require the server to be rebooted in order for the changes to take effect. Each of the file's command sets are contained within a
<DIRECTORY directory_path> ... </DIRECTORY>
command. Each
<DIRECTORY directory_path > ... </DIRECTORY>
command affects all the files and subdirectories for a single directory tree, defined by the directory_path. Remember that a directory tree is just a starting path to a directory and all the directories below that directory.
The srm.conf file controls the server after it has started up. Inside this file, you will find the path to the document root and an alias command telling the server where to hunt for CGI scripts. The srm.conf file is used to enable Server Side Include commands and to tell the server about new file extensions that aren't part of the basic MIME types. One file type you are particularly interested in is the x-parsed-html-type file type, which defines for the server in which files to look for the SSI commands.
This brief introduction to your configuration files should just whet your appetite for the many things you can learn by being aware of and understanding how your server configuration files work.
You normally will be working in a directory tree called the document root. The document root is the area where you put your HTML files for access by your Web clients. This probably will be some subdirectory of your user account. On my server, the document root for each user account is public-web. User accounts who want to create public Web pages must place those Web pages in the public-web subdirectory below their home directory. You can create as many subdirectories below the public-web directory as you want. Any subdirectory below the public-web directory is part of the document root tree.
How do you find out what the document root is? It is easy, even if you aren't a privileged user. Just install either the HTML Print Environment Variables program or Mail Environment Variables program (described in Chapter 6) and you will see right away what the document root directories are on your server. To find out what the server root is, you need to contact your Web Master or System Administrator.
After you figure out where to put your HTML, Server Side Include commands, and CGI files, the next thing you need to learn is how to enable them so they can be used by the WWW server.
When you create a file, the file is given a default protection mask set up by one of your login files. This normally is done by a command called umask. Before you learn how to use the umask command, you should learn what file protection masks are.
File protections also are referred to as file permissions. The file permissions tell the server who has access to your file and whether the file is a simple text file or an executable program. There are three main types of files: directories, text files, and executable files. Because you will be using Perl as your scripting language, your executable CGI programs will be both text and executable files. Directory files are special text files that are executable by the server. These files contain special directives to the server describing to the server where a group of files is located.
Each of these file types has three sets of permissions. The permissions are Read, Write, and Execute. The Read permission allows the file to be opened for reading but it cannot be modified. The Write permission allows the file to be modified but not opened for reading. The Execute permission is used both to allow program execution and directory listings. If anyone, including yourself, is going to be able to get a listing or move to a directory, the Execute permission on the directory file must be set. The Execute permission also must be set for any program you want the server to run for you. Regardless of the file extension or the contents of a file, if the Execute permission is not set, the server will not try to run or execute the file when the file is called.
This is probably one of the most common reasons for CGI programs not working the first time. If you are using an interpretive language like Perl, you never run a compile and link command, so the system doesn't automatically change the file permissions to Execute. If you write a perfectly good Perl program and then try and run it from the command line, you might get an error message like Permission denied. If you test out your CGI program from you Web browser, however, you are likely to get an error like the one shown in Figure 1.1—an Internet file error with a status code of 403. This error code seems kind of ominous the first time you see it, and it really doesn't help you very much in figuring out what the problem is.
Figure 1.1. The Forbidden error message.
Remember that there are three types of file permissions: Read, Write, and Execute. Each of these file permissions is applied at three separate access levels. These access levels define who can see your files based on their user name and group name.
When you create a file, it gets created with your user name and your group name as the owner and group name of the file, respectively. The file's Read, Write, and Execute permissions are set for the owner, the group, and other (sometimes referred to as world). This is very important because your Web page is likely to be accessed by anybody in the world. Usually, your Web server will run as user nobody. This means that when your CGI program is executed or your Web page is opened for reading a process with a group name different than the group name you belong to, someone else will be accessing your files. You must set your file-access permissions to allow your Web server access to your files. This usually means setting the Read and Execute privileges for the world or other group. Figure 1.2 shows a listing of the files in one of my business directories. You can see that most of the files have rw privileges for the owner and only Read privileges for everyone else. Notice that the owner is yawp (that's my personal user name) and the group is bizaccnt. You can see that directories start with a d, as in the drwxr-sr-x permissions set. The d is set automatically when you use the mkdir command.
Figure 1.2. A directory listing showing file permissions.
In order for your Web page to be opened by anyone on the Net, it must be readable by anyone in the world. In order for your CGI program to be run by anyone on the Net, it must be executable by your Internet server. Therefore, you must set the permissions so that the server can read or execute your files, which usually means making your CGI programs world executable. You set your file permissions by using a command called chmod (change file mode). The chmod command accepts two parameters. The first parameter is the permission mask. The second parameter is the file for which you want to change permissions. Only the owner of a file can change the file's permissions mask.
The permissions mask is a three-digit number; each digit of the number defines the permission for a different user of the file. The first digit defines the permissions for the owner. The second digit defines the permissions for the group. The third digit defines the permissions for everyone else, usually referred to as the world or other, as in other groups. Each digit works the same for each group of users: the owner, group, and world. What you set for one digit has no effect on the other two digits. Each digit is made up of the three Read, Write, and Execute permissions. The Read permission value is 4, the Write permission value is 2, and the Execute permission is 1. You add these three numbers together to get the permissions for a file. If you want a file to only be readable and not writable or executable, set its permission to 4. This works the same for Write and Execute. Executable only files have a permission of 1. If you want a file to have Read and Write permissions, add the Read and Write values together (4+2) and you get 6, the permissions setting for Read and Write. If you want the file to be Read, Write, and Execute, use the value 7, derived from adding the three permissions (4+2+1). Do this for each of the three permission groups and you get a valid chmod mask.
Suppose that you want your file to have Read, Write, and Execute permissions (4+2+1) for yourself; Read and Execute (4+1) for your group; and Execute (1) only for everyone else. You would set the file permissions to 751, using this command:
chmod 751 (filename)
Table 1.1 shows several examples of setting file permissions.
Command |
Meaning |
|
chmod 777 filename |
The file is available for Read, Write, and Execute for the owner, group, and world. |
|
chmod 755 filename |
The file is available for Read, Write, and Execute for the owner; and Read and Execute only for the group and world. |
|
chmod 644 filename |
The file is available for Read and Write for the owner, and Read only for the group and world. |
|
chmod 666 filename |
The file is available for Read and Write for the owner, group, and world. I wonder if the 666 number is just a coincidence. Anybody can create havoc with your files with this wide-open permission mask. |
Tip: If you want the world to be able to use files in a directory, but only if they know exactly what files they want, you can set the directory permission to Execute only. This means that intruders cannot do wild-card directory listings to see what type of files you have in a directory. But if someone knows what type of file she wants, she still can access that file by requesting it with a fully qualified name (no wild cards allowed).
When you started this section, you were introduced to a command called umask, which sets the default file-creation permissions. You can have your umask set the default permission for your files by adding the umask command to your .login file. The umask command works inversely to the chmod command. The permissions mask it uses actually subtracts that permission when the file is created. Thus, umask stands for unmask. The default umask is 0 , which means that all your files are created so that the owner, group, and world can read and write to your files and all your directories also can be read and written to. A very common umask is 022. This umask removes the Write privilege from all the files you create. Every file can be read and all directories are executable by anyone. Only you can change the contents of files or write new files to your directories, however.
Now that you have a feel for how to move around the directories on your server, let's back up for a moment and talk about the available servers on the Net. This book definitely leans toward the Unix world, but only because that is where all the action is right now. Because everything on the Net is changing so fast, moving out of the mainstream into a quieter world that may be more comfortable is a major risk. The problems of today will be solved or worked around tomorrow, and if your server isn't able to stay up with the rush, you will find yourself left behind. "What is your point?" you might ask? The comfort factor gained from working in a familiar environment might not be worth the risk of being left behind. When choosing one of the servers outlined in the next sections, make one of your selection criteria the server's capability to keep pace with the changes on the Net.
Servers are available right now for Windows 3.1, Windows NT, and Windows 95. The Windows 3.1 server is available at
http://www.city.net/win-httpd/
This server is written by Robert Denny, who is also the author of the Windows NT and Windows 95 servers known as Website. The Website server is available at
http://www.ora.com/gnn/bus/ora/news/c.website.html
Each of these servers implements all or almost all of the major features of the NCSA httpd 1.3 server for Unix. They are easy to configure and the Windows NT/95 version uses a graphical user interface for configuration. These servers have hooks to allow the server to work with other Microsoft products as well. Because they provide a familiar environment for many MS-based PC users, they might seem like a good system to choose.
If you choose an MS-based server, however, you definitely will be swimming out of the mainstream. The two most popular Web servers on the Net are the original Web server CERN, created by the European High Energy Physics Lab Group, and the National Center for Super Computing Applications, NCSA httpd Web server. The CERN server was the first Web server—the starting point for the World Wide Web. It still is the test site for many of the experimental features being tried each day. Even though the CERN Web server is no longer the most popular server on the Net, it has one feature that you cannot get anywhere else right now. If you are trying to create a really secure site and you want to use a Web server as the proxy host, the CERN server is the way to go.
The CERN server enables you to implement a firewall to protect your network from intruders, while still allowing Internet WWW access from inside the firewall. Firewalls are great security barriers for preventing unwanted guests from getting into your secure network. A firewall typically works by allowing only a select set of trusted machines access to the network. A machine called a proxy is used to screen incoming and outgoing connections.
The problem with this setup is that it usually prevents machines on the inside of the firewall from accessing the WWW. However, if you set up the CERN server as a proxy server, your Web browser on the inside of the firewall can request WWW documents from the CERN proxy, and the CERN proxy forwards the request to the correct domain. When the domain server responds with the requested Web page, the CERN proxy passes the response to your browser. This lets your internal Net see the outside WWW and still provides the security of a firewall. As you would expect, this does slow down your access to Internet documents somewhat. Passing the information through the intermediary proxy server adds overhead and takes more time. If you don't need a proxy server, the most popular server on the Net by far is the NCSA server called httpd.
You can learn more about the CERN server at
http://www.w3.org/hypertext/www/daemon/overview.html
The NCSA server usually is referred to by its version number. The current version of this server is the NCSA httpd 1.4 server. The 1.4 version of the NCSA server provides excellent execution speeds—sometimes equivalent to the commercial servers on the Net. The NCSA server provides support for Server Side Include commands (something the CERN server does not provide) and security based on a general directory tree, per-directory access, or remote IP addresses. Because this server is by far the most popular server on the Net and most of its features are available on the other servers on the Net, this book uses the NCSA server as the basis for most of the examples and descriptions. You can find more information about the NCSA httpd server at
http://hoohoo.ncsa.uiuc.edu/docs/overview.html
Finally, a brief mention of the commercial Netscape server. This server comes in two versions: the Netscape Communications server and the Secure Communication Netscape Commerce server. Both servers provide excellent speed and support for their users. The Netscape Commerce server is designed for secure commerce over the Internet. The Netscape Commerce server currently only provides secure communication with Netscape's own WWW browser, the Netscape Navigator, however. You can get more information about the Netscape servers at
http://home.netscape.com/
For the most part, I will be dealing with the NCSA httpd server. This is the server that is setting the standard for the Net—if you can call a target moving at light speed a standard. But I would rather try to stay with this fast-moving target than get left behind one of the most exciting rides of the decade.
Probably the two most common questions about CGI programming are "What is CGI programming?" and "Why is this programming paradigm so different?" The first question is the harder question to answer and certainly is the combination of all the pages in this book, but there is a short answer. CGI programming is writing applications that act as interface or gateway programs between the client browser, Web server, and a traditional programming application.
The second question, "Why is CGI programming different from other programming?" requires a longer answer. The answer really needs to be broken up into three parts. Each part describes a different section of the CGI program's environment, and it is the environment that the CGI program operates under that makes it so different from other programming paradigms. First, a CGI program must be extra concerned about security. Next, the CGI programmer must understand how data is passed to other programs and how it is returned. And finally, the CGI programmer must learn how to develop software in an environment where your program has no built-in mechanisms to enable it to remember what it did last.
Why does your CGI program have to be extra concerned about security? Unfortunately, your main concern is hackers. Your CGI programs operate in a very insecure environment. By their nature, they must be usable by anyone in the world. Also by their nature, they can be executed at any time of the day. And finally, they can be run over and over again by people looking for security holes in your code. Because the Net is a place where anyone and everyone has the freedom to search, play, and explore to their heart's content, your programs are bound to be tested eventually by someone with at least an overabundance of curiosity. This means that you must spend extra time thinking about how your program could be broken by a hacker. In addition, because many applications are written in an interpretive language like Perl, your program source code is easier to access. If a hacker can get at your source code, your code is at much greater risk.
The way data is sent back and forth across the Internet is one of the most unique aspects of CGI programming. Gathering data and decoding data are the subject of Chapters 4 and 5, respectively, but a brief introduction is warranted. Your CGI program cannot be designed without first understanding how data is built using the HTML hypertext link or the HTML Form fields. Both mechanisms create a unique environment in which data is encoded and passed based on both user input and statically defined data structures. When you design your CGI program, you first must design the user input format. This format is fixed in two data-passing mechanisms: the Get and Post methods. Both these methods use HTTP headers to communicate with your CGI program and to send your CGI program data. As you design your CGI program, you must be aware of the limitations of both these methods.
In addition, your CGI programs must be able to deal with the multiple input engines on the Internet, which have an impact on the format of the data your CGI program can return. Your CGI program can be called from all types of browsers—from the text-only Lynx program; the HTML 1.0 capable browsers; or the browsers like Netscape that include data, such as the cookie, that isn't even included in the HTTP specification. It is up to you to design your CGI program to deal with this multiplicity of client/browsers! Each will be sending different information to your CGI program, describing itself and its capabilities in the HTTP request headers discussed in Chapter 2.
Once you have the data from these myriad sources, your CGI program must be able to figure out what to do with it. The data passed to your CGI program is encoded so as to not conflict with the existing MIME protocols of the Internet. You will learn about decoding data in Chapter 5. After your CGI program has decoded the data, it must decide how to return information to the calling program. Because not all browsers are created equal, your CGI program may want to return different information based on the browser software calling it. You will learn how to do this in the last part of Chapter 2.
The implementation of the HTTP stateless protocol has a profound effect on how you design your CGI programs. Each new action is performed without any knowledge of previous actions, and multiple copies of your CGI program can be executing at the same time. This has a dramatic effect on how your program accesses files and data. Database programming alone can be complicated, but if you add parallel processing on top of it, you have a much more complicated problem.
Traditional programming paradigms use sequential logic to solve problems. The data you set up 100 lines of code ago is expected to be available when you need it to pass to a subroutine or write to a file. Usually when you run one program in a traditional environment, it gets to run to completion, without fear of another copy of itself modifying the same data.
Neither of these conditions is true for your CGI programs. If you are building a multipaged site where the information on one page can affect the actions of another page, you have a complication for which you must design. Unless you take special steps, what happened on Web page 12 is not available the next time Web page 12 or any other page in your site is accessed. Each new Web page access creates a brand new call to your CGI program. This means that your CGI program has to take special measures to keep track of what happened last time. One common means is for your CGI program to save information from the last event into a file. That method still has limitations, however, because your program can be executed simultaneously by several clients. You need to know which client is calling you.
To get around these special problems, the HTML form input type of hidden was created. The hidden input type enables your program to return data in the called Web pages that aren't displayed to the Web client. When the client calls the next Web page on your site, the hidden input type is returned as data to your CGI program. This way, your CGI program has a chance to remember what happened last time.
This approach has at least one major problem. Hidden data is visible as soon as your Web client uses the View Source button on his browser. This means that he can change the data returned to your CGI program.
To complicate things even further, because your CGI program can be called from multiple browsers simultaneously, your program can be modifying a file at the same time another copy of the same program is modifying the same file. Unless you take special precautions to deal with this situation, some of your data is going to get lost. In the case where two programs have the same file open, the program that closes the file last wins! The data saved by the earlier program is lost, overwritten by the changes made by the program that closed the file last. How do you solve this problem? You have to design a special database handle that locks the file for writing whenever any code in your CGI program has the file out for updating.
These are just the most obvious problems. It is your job as a CGI programmer to think about these possible problems and to come up with effective solutions.
One solution to the hidden field view source problem is the experimental HTTP header called a cookie. This cookie acts something like a hidden field, but it cannot be accessed by the user. Just your CGI program and the browser can see this field. This gives you a second and more secure means of keeping track of what is happening at your Web site. The HTTP cookie is discussed in Chapters 6 and 7.
I suspect that you would prefer to just get your first CGI program working. If you can prevent the common CGI errors described in this section, you will be well on your way to getting your first CGI program working. What happens when you try to run your first CGI program and you get a Server Error (500) message back, such as the one shown in Figure 1.3?
Figure 1.3. The Server Error message.
It seems like kind of an ominous error message. Drop everything and write your System Administrator a message describing exactly what you did to break the server. And what abut the Forbidden (403) error message in Figure 1.1? Is the System Administrator going to cut off your programming privileges? DOES ANYONE KNOW? Can you just not tell anyone and it will go AWAY??!! Well, yes and no.
First of all, I suspect that you realize all these error messages are generated automatically by your Web server, so nobody "knows" and, in most cases, nobody cares, but the error doesn't go away. Your Web server logs into an error log file every error that is sees. This file is a marvelous source for figuring out what went wrong with your program. The error log file your server uses is probably in the server root document tree described earlier.
Usually, you will have read-only privileges for the files on the server root. This means that you can read what's in the error log files, but not change them. The error log files also are used by your System Administrator to watch for potential security risks on her server because each access to the system is logged in these files.
There is one way to keep your programs from showing up in the error log files. Never make any mistakes! Because I've never been able to be successful with that advice, I've followed the more practical advice of always (well, okay, almost always) executing my CGI programs from the command line before trying to test them from my Web browser. Just enter the file name of your program from the prompt. If everything is okay, your CGI program executes as expected and you should see the HTML your CGI program generated output to your screen.
If you have an error, usually Perl is very good about helping you find what is wrong. Perl tells you the line where the error is located and suggests what it thinks the problem might be. I suggest fixing one or two errors at a time and then retrying your program from the command line. Quite often, one error will contribute and create lots of other errors. That's why I suggest just fixing a couple of bugs at a time.
One of the first things you are likely to forget is to tell the system which language to run your script under. Setting the file extension to .pl doesn't do it. The thing that tells the system how to run your CGI program is the first line of a Perl script. The first line should look something like this:
#! /usr/local/bin/perl
The line must align flush with the left margin, and the path to the Perl interprets must be correct. If you don't know where Perl is on your server, the following exercise will help you figure it out.
One way to figure out where stuff is on you system is to use the whereis command. From the command line, type > whereis perl. The system will search for the command (perl) in all the normal system directories where commands can be found and return to you the directory in which the Perl interpreter resides.
If this doesn't work for you, try typing the which command type > which perl from the command line. The which command searches all the paths in your path variable and returns the first match for the command.
If neither of these methods work, try using the find command. Change directories to one of the top-level directories (starting at /usr/local, for example).
At the prompt > cd /usr/local, type > find . -name perl -print. This command searches all the directories under the current directory, looking for a file that matches the file in the -name switch end.
After you tell the system which interpreter to run and where it is, what next? Well, the next most common mistake is forgetting to set the file permissions correctly. Is your program executable? Even if everything else about the program is right, if you don't tell the server that your program is executable, it will never work! You might know it's a program, but you're NOT supposed to keep it a secret from the server.
Enter ls -l at the command line. If you see the following message, you forgot to change the file permissions to executable:
rw-rw-rw- program.name
Don't be too chagrined by this; I wouldn't mention it if it didn't happen all the time. It's really frustrating when you've been doing this for 10 years and you still forget to set the file permissions correctly. What's embarrassing though is asking someone why your program doesn't work, and the first thing she checks are your file permissions. The look you get from your Web guru when your file isn't executable just makes you want to go hide under a rock. Don't do this one to yourself; always check your file permission before asking someone else what is wrong with your program. Then set your program's file permissions to something reasonable like
> chmod 755 programe.name
Tip: If you have a lot of output from your program and want to save it to a file so that you can study it a little easier, from the command line, pipe the output from your program into a file by using the redirection symbol (>). Enter your program like this:
If you've done all of this and you now are testing from your Web browser and you still are getting one of those ominous server error messages, check for this common mistake: Make sure that your CGI program is printing a valid Content-Type response header and that the last response header your CGI program prints is two newline (\n) characters immediately after the response header.
Most of your CGI programs can use a print line just like this:
print "content-type: text/html\n\n";
The \n at the end of the HTTP response header prints a newline character. The server knows that your CGI program has sent its last response header when it finds a blank line after an HTTP response header. After that blank line, it is expecting to find the content type your program described in the Content-Type response header.
There is still one bug that usually bites the more experienced programmers more often than the inexperienced folks. The file-name extension must be correct. Us experienced (old dudes) guys and gals know that the file-name extensions don't really mean anything, so we are more likely to ignore the file-naming convention of filename.cgi for CGI programs. This is a big mistake! The Web server really does use that file-name extension to determine what it is supposed to do with the file requested by the browser. So use the correct file extension! It's probably .cgi, but check the srm.conf file found under the server root directory in the configuration directory because it has the correct file extension. Look for something like
AddType application/x-httpd-cgi .cgi
You will save lots of debugging time if you always check these things first:
You have covered a lot of territory in this chapter, and a lot of it still might seem confusing. Don't worry—the purpose of this chapter is to get you thinking about the concepts of CGI programming. Explaining these concepts in detail takes the remainder of the book. In this chapter, you learned that CGI programming is a lot more than just another programming language. It is really a programming paradigm—something that defines how you program and not what you program.
CGI programming is not a single language or application; it is making applications work in that wonderful WWW environment. In this chapter, you learned about the three main keys to your CGI program: HTML, HTTP, and your server. Each of these impacts how your program is structured to satisfy the needs of each application. You also learned about the structure of your server and where to find the different parts of your server directories.
Finally, you learned some of the common CGI programming mistakes to avoid as you begin to build your own CGI program applications.
Q: Where should you put your CGI programs?
A: Ultimately, your System Administrator or Web Master has control over where you can install your CGI program. If you are on an NCSA server, you can create and run your CGI program from any directory. However, it's usually a good idea to keep your CGI programs in a common directory. That way, you can find them when you need to modify one. A lot of systems create a single directory called the cgi-bin directory. If your server is set up this way, you might need to have your Web Master install each CGI program you create. However, because this is such a time-consuming process, usually you can be added to the group name that has privileges to write into the cgi-bin directory. Check with your server's System Administrator.
Q: Are CGI programs only interface programs?
A: There are absolutely no restrictions on what your CGI program can be. The only limitation on a CGI program is the requirement that it must understand the HTTP request/response headers, and it usually will be dealing with HTML in some manner. Frequently, CGI applications are small, quickly built programs that perform some simple task. As the Web grows more sophisticated, however, CGI applications will become larger and more complex.
Q: What is per-directory access?
A: Each of the directories within your public-directory tree can be password protected. The access.conf file defines the overall structure of directory access, but you can add a similar file (usually called .htaccess) that creates special directory protection for the directory tree in which it is installed. You will learn more about per-directory access in Chapter 14.
