Most browsers now support the "Host" field in the HTTP header. The Host field specifies the name of the server that the user has requested the document from. In most cases, this field will be the domain name of your server, or its IP address if the requested URL was specified by IP address.
This command can be very helpful in two situations. First, if your server has multiple domain names associated with it, the Host field is an excellent way to find out which domain name was requested by the client. For details on supporting multiple domains with the Host field, see the SHOW_HOST section below. Second, if you serve your Web site from multiple servers, you can create pages that are sensitive to the server they are being served from.
The INSERT_HOST command simply inserts the domain name (or IP address) that the URL specified when requesting the page.
Note that some older browsers do not send the Host field in the HTTP header. When pages are accessed by these clients, the INSERT_HOST command will not insert anything.
The HIDE_HOST command allows you to hide portions of a page based upon the domain name (or IP address) as specified in the URL used to access the page. Text after the command will be hidden if any of the specified "host" parameters appear anywhere within the requested Host. The comparison is not case sensitive.
HIDE_HOST also allows you to use an asterisk (*) as a wildcard to specify one unknown character. For example:
<HIDE_HOST b*rt>
would match a host of "bart", "aborted" or any other host name containing a single character between "b" and "rt".
The SHOW_HOST command allows you to hide portions of a page based upon the domain name (or IP address) as specified in the URL used to access the page. Text after the command will be shown if any of the specified "host" parameters appear anywhere within the requested Host. The comparison is not case sensitive.
For example, assume your Web server is used to serve pages for three different domains: "www.greg.com", "www.peter.com", and "www.bobby.com". Each of these domain names points to the same address, which is the address of your server, and each web site is served from a sub-folder within the root folder (called "greg", "peter", and "bobby"). The default page for your server, at the root level, could be:
<HIDE><SHOW_HOST "www.greg.com"> <REDIRECT "http://www.greg.com/greg/"> <HIDE><SHOW_HOST "www.peter.com"> <REDIRECT "http://www.peter.com/peter/"> <HIDE><SHOW_HOST "www.bobby.com"> <REDIRECT "http://www.bobby.com/bobby/"> <SHOW> The requested host is unknown. Please select from the following list of domains...
In this case, the SHOW_HOST command is used to select a REDIRECT command which sends the client to the correct default page for the domain. If the Host isn't recognized (most likely due to lack of support for the field in the browser), a default page is shown to the user so they can select the domain they are looking for.
Note that the MACRO command could be used in place of the REDIRECT commands to reduce processing, although the relative paths for URLs in the individual Home Pages could be confused. See the MACRO command for additional information.
The INSERT_HEADER command inserts the entire HTTP request header into the HTML document.
It is rare that you would want to display the HTTP header of an incoming request back to the user. There is, however, at least one very good reason for doing this. As new browsers become available with new HTTP features, you will want to add support for them into your documents. By using the "INSERT_HEADER" command in a few test pages, you can see exactly what information is being sent by various browsers. This makes it easy to try out different browsers on your pages to see just what the browser is sending so that you know what to look for when you are developing new pages.
The usage of this command is simple, and looks like this:
<PRE> <INSERT_HEADER> </PRE>
We use the <PRE> HTML tag because the HTTP headers are best viewed in this mode so that line breaks are displayed on the browser correctly. These formatting tags are not required, but the header will most likely be difficult to read without them.
The HIDE_HEADER command will hide HTML following it if the specified text is contained anywhere in the header of the HTTP request. The comparison ignores capitalization.
The Web is based upon the HTTP protocol, which is constantly evolving. Every request that comes into your server is in the form of an HTTP header request that may or may not have additional information that is not captured by some other NetCloak command. The "Header" commands give you the ability to look into the header even when there is no direct NetCloak command for checking a particular piece of information. Let's look at the "Show" version of these commands for an easier explanation of how this works.
The SHOW_HEADER command will "turn the stream on" when the specified text appears anywhere in the HTTP header of the incoming request. The match is not case sensitive. The specified text can be an acceptable file type or language, the requested URL, or any other information included in incoming HTTP requests.
For example, many browsers support a field which specifies the languages they can accept and display. If you would like to selectively serve files in various languages, you can use this field to automatically select text to be displayed. For example, to display a section of text in French to browsers that support it, you might use:
<HIDE><SHOW_HEADER "Accept-Language: fr"> Bon jour!
<SHOW><HIDE_HEADER "Accept-Language: fr"> Hello!
In this example, if the browser sends the appropriate field, the French text will automatically be displayed by NetCloak.
The HTTP header can also be used to help determine file types the browser can accept and more. Essentially any information that the browser sends to the server is yours for the asking.
This command will insert the URL of the last page visited into your document at the specified point. The URL will be exactly as reported by the browser, including server name and path and search arguments.
Here is an example:
You just came from: <INSERT_REFERRER>.
If you want, you can:
<A HREF="<INSERT_REFERRER>">Go Back!</A>
Note that this command will not work if the browser does not report the last page the user visited. In general, the referrer will be reported when the user clicks on a hypertext link to access the page. However, if the user manually enters the URL, the browser will not send the referrer, since the page the user was on when they entered the URL doesn't actually reference the page being served.
Since "REFERRER" is commonly misspelled "REFERER" (including early versions of the Internet specification for HTTP), NetCloak also supports that spelling for all REFERRER commands.
This command hides the HTML text from users based on the last HTML document they accessed (the document that "referred" the user to this one). This allows you to tailor your document depending on where the user has been.
Specifying "referrers" is less restrictive than some of the other NetCloak commands, because the referrer is returned by the web browser and is subject to vary slightly. Specifically, the referrers you specify are not case sensitive, and they can appear anywhere in the URL of the last document the user was on.
Note that if a user types an URL into their Web browser to get to your page, there will be no referrer, since the page the user was on when they entered the URL doesn't actually reference the page being served.
Some browsers may not support the Referrer field, and if the browser doesn't send it, NetCloak won't be able to use it. You can check for the existence of the field by using the command alone, with no referrers specified. In other words, <HIDE_REFERRER> will hide the text following it whenever a referrer is known.
HTML text that follows this command will be visible to users when the most recent document they visited in their web browser contains one of the specified "referrer" parameters.
For example, if your site was selected as the "Cool Site of the Day", you might add this to your Home Page:
<HIDE> <SHOW_REFERRER Cool> Welcome, Cool Site of the Day users!
<SHOW>
This would show the text only to people who had just come from a page with the word "Cool" in the URL.
Use SHOW_REFERRER, with no referrers specified, to make sure the browser supports the referrer field.
<HIDE><SHOW_REFERRER>You just came from <INSERT_REFERRER>!
<SHOW><HIDE_REFERRER> Sorry, your browser does not support referrers, or you entered the
URL for this page manually.
<SHOW>
The INSERT_THISURL command ("This URL"), which accepts no parameters, will cause the file path and document name of the current page to be inserted into the page. For example, if the URL for a page is:
http://www.web.com/bob/page.html$path?qry
the INSERT_THISURL command would insert:
/bob/page.html
With servers that support virtual hosting of multiple Web sites, the path will be relative to the root folder for the requested site.
The INSERT_THISURL could be used in hidden fields on HTML forms to tell a forms processing CGI the URL of the entry form. It might also be used to create a "reload page" link, or to give the user additional feedback on how your site is set-up.
If any of the specified parameters are contained in the file path or filename for the page where the command appears, then the HIDE_THISURL command will "turn off" the HTML stream.
With the THISURL commands, NetCloak allows you to essentially create "self-aware" pages. By using SHOW_THISURL and HIDE_THISURL, you can create pages that change when they are moved from one folder (or server) to another. These commands can also be used in your macro file to create macros that vary depending on the page they are used on. In short, they allow you to create more portable, re-usable HTML.
This is the SHOW version of the HIDE_THISURL command above.
Here's an example:
<HIDE><SHOW_THISURL TestFolder> This page is in the folder 'TestFolder', and is still being tested.
When the file is moved into its final place on my site, this test
message will disappear.
<SHOW><HIDE_THISURL TestFolder> This page is now in production use...
<SHOW>
When a non-existant page is requested and the error page is served, "THISURL" is the requested (non-existant) URL. This makes the SHOW_THISURL command useful for making a "smart" error page. If users have a typo in their URL, the error page can automatically redirect them to the page they were looking for.
<HIDE><SHOW_THISURL faq> <REDIRECT "http://my.server/questions/faqlist.html">
You can use a similar technique to automatically redirect users accessing obsolete pages by putting the URL of the obsolete page in the SHOW_THISURL command and the new URL in the REDIRECT command.
With the example above on your error page, users who request a page that does not exist but contains "FAQ" will be automatically redirected to the FAQ List page.
In an HTTP URL, there are three parts of the URL that come after the server name. The first is a path to the requested file. The second part, called the "path arguments," follow the name of the requested file. In MacOS web servers, the path arguments are separated from the filename by a dollar sign character ('$'). Following the path arguments is the third part of the URL, called the "query arguments" or "search arguments." The search arguments are separated from the path arguments (or the filename, if there are no path arguments) by a question mark character ('?'). Both path arguments and search arguments are URL-encoded by browsers (in other words, all non-alphanumeric characters are turned into a percent sign '%' followed by the two-digit hexadecimal ASCII code for the character). Here is an example URL containing both path arguments and search arguments:
http://www.maxum.com/netcloak/sample.html$path%20Args?search%20Args
In this example the path arguments consist of the text "path Args"(which has been URL-encoded, turning the space character into "%20"), and the search arguments consist of the URL-encoded text "search Args". It is important not to confuse path arguments with the requested file path. In the example above, the file path is "/netcloak/sample.html".
Path and search arguments can be used to send additional data to NetCloak to be used in processing cloaked pages, simply by appending them to the URL of any cloaked page, then using the following commands within the cloaked pages.
The INSERT_PATH command inserts the path arguments for the requested URL into the cloaked page. If there are no path arguments in the requested URL, this command does nothing.
If the file "/netcloak/sample.html" in the above example contained the following HTML:
This text follows the dollar sign in the requested URL: <INSERT_PATH>
The page seen at the browser would contain:
This text follows the dollar sign in the requested URL: path args
Important Note: If you are using the NetCloak ACGI by calling it directly in your URLs (no "CLOAK" action defined), like this: "http://www.yourdomain.com/NetCloak.acgi$/somefile.html", then the path arguments are used to send the URL of the cloaked file to NetCloak, and are not available for you to use for other purposes. In this case, the INSERT_PATH command gives you the path to the cloaked file.
The HIDE_PATH command turns "off" the flow of text in the page if the path arguments in the requested URL contain any of the parameters listed in the command. See SHOW_PATH for an example where this could be useful.
The SHOW_PATH command turns "on" the flow of text in the page if the path arguments in the requested URL contain any of the parameters listed in the command.
The SHOW_PATH command can be useful when you want one page to contain all the logic for some interactive process, like an online questionnaire or game. Rather than spreading the logic for the interaction across multiple cloaked pages, you can put it all in one cloaked page, and use path arguments to determine which part of the process to send to the browser. An excellent example of this is the "regkillerapp.html" example included in the NetCloak package.
In the "regkillerapp.html" example, a form is submitted and processed by the same page, based on the path arguement. The form command looks like this:
<FORM METHOD=POST ACTION="regkillerapp.html$step2">
Notice the path arguement specified, "Step2". Later in the page, the form is processed by enabling a section of HTML with the NetCloak command:
<HIDE><SHOW_PATH "step2">
Similarly, the form itself must be hidden once it has been submitted, this is done like this:
<HIDE><SHOW_PATH "step1" "">
This effectively shows the form when either the first step is specified by the path arguement, or when no path arguement is given (as would be the case when the form is linked to from another page).
The INSERT_SEARCH command inserts the search arguments for the requested URL into the cloaked page. If there are no search arguments in the requested URL, this command does nothing.
If the file "/netcloak/sample.html" in the INSERT_PATH example contained the following HTML:
This text follows the question mark in the requested URL: <INSERT_SEARCH>
Then the browser would see:
This text follows the question mark in the requested URL: search args
Important Note: When an HTML form specifies a method of GET instead of POST, the data in the form fields are appended to the URL specified in the action of the form, as search arguments. In this case, the form data is not available to NetCloak as user variables. Instead, the SEARCH commands are used. This may be useful in situations where the form data is very short, or when you already have a set of user variables generated from a previous form, and you don't want to lose those variables yet. See the section "Commands: User Variables" for more information.
The HIDE_SEARCH command turns "off" the flow of text in the page if the search arguments in the requested URL contain any of the parameters listed in the command. See SHOW_SEARCH for an example where this could be useful.
The SHOW_SEARCH command turns "on" the flow of text in the page if the search arguments in the requested URL contain any of the parameters listed in the command.
Search arguments typically are used (as their name implies) to send text to be searched for, either in a database or in the files of the web site. But with NetCloak, they can be used just like path arguments, for just about any purpose. Here's an example where we use NetCloak commands to add better feedback and a consistent look-and-feel to the results returned from a simple file-searching CGI:
<HTML> <HEAD><TITLE>Search Results</TITLE></HEAD> <BODY> <MACRO Header> <H1>Search Results</H1> Here are all the matches found for the phrase "<INSERT_SEARCH>":<P> <HR> <EXEC_CGI "SimpleSearch.cgi"> <!-- The CGI returns raw results --> <HR> <MACRO Footer> </BODY> </HTML>
Copyright © 1996-7 Maxum Development Corporation
820 South Bartlett Road - Suite 104
Streamwood, IL 60107
http://www.maxum.com/