Builds into Apache a simple keyowrd based indexing mechanism.
After configuring conf/mime.types with a filename extension to be treated as a httpd index file, requesting a URL of this type will return an ISINDEX response. Subsequent requests to the same document with arguments causes Apache to search for those arguments as keywords describing URLs.
Add the httpd/auto-index MIME type, with your chosen filename extension. We recommend using indx, but you can choose any extension, e.g.
httpd/auto-index indx
The format for an index file is very simple.
The first line of the index file will be used to generate the
<TITLE> message, and a <H1> heading for the HTML output.
Choose something simple, which explains the purpose of the index e.g.
Keyword searchable index for Widget INC resources
Subsequent lines in the index file will,
Keyword searchable index for Widget INC resources this line doesn't start with "K: " "U: " or "D: " so it will be ignored K: about Widget U: /help/about_widget_inc.html D: Information about the Widget INC company, including contact addresses. K: logo tradmark U: /images/widget.gif D: Widget INC company's logo and trademark, a 59k 256 color GIF. another line to be ignored - might be a comment K: sales buy orders purchase sell U: /cgi-bin/sales/order.pl D: An order form for Widget INC products.
n.b.For best results, "U: " lines should appear before "D: " lines.
The mime type httpd/auto-index is intercepted by Apache. A new searching routine then scans the index file for case insensitive matches on keyword lines. In the event of a match, a descriptive list item (DT/DD) is output.
Thanks to the simplistic model, there's minimal overhead in supporting these index files.
Here's some example output based on the index file above. The keywords About logo yield...
Keyword searchable index for Widget INC resources
- about
- Information about the Widget INC company, including contact addresses.
- logo
- Widget INC company's logo and trademark, a 59k 256 color GIF.