In NCSA httpd 1.0a2, you could do this, although not quite as you are
asking. What you would probably do is have a script (which is your filter)
in /htbin, let's call it foo. HREF's to the TEI documents could then be
something like /htbin/foo?/the/document/name.blort.
In later releases, we're going to change the script interface such that when
looking in the scripts directory, if it finds an executable early, it will
send the rest of the path as an argument to the script. In this
architecture, your documents would be referenced by something like:
/htbin/foo/the/document/name.blort
And the script execution would be transparent.
* Why do I want to do this? because I expect to have several zigaflops of
* TEI-conformant sgml text here is why, and I really don't want to have
* to keep copies of them all online in html as well. Writing a filter
* to translate TEI markup into html is well within the bounds of even
* my C-competence.
*
* thank you
*
* Euro-Lou
*
*
*
*/
Hope this helps
--Rob