I'm trying to figure out how to use swish to index a MySQL database. The 
data in the database is used to populate a single PHP page . . . call it 
http://www.mysite.com/data.php.

I think I understand how to use prog to pull the data out of the 
database and feed it to swish for indexing. My question is, how does 
swish relate the pages created on the fly for indexing to the the actual 
website page, i.e, ~/data.php?

Is it done with the Path-Name directive? Will something like the 
following do the trick?

# Output the document (to swish)
    print <<EOF;
    Path-Name: 'http://www.mysite.com/data.php'
    Content-Length: $size
    Last-Mtime: $mtime
    Document-Type: HTML*

-- John