> This gets a little bit tricky if there are several conditional
> effect SQL queries (eg: query1 determines what query2 is), but
> should be doable.
If the mechanism is general, then the event at the end of the first query
should be able to set up an event for the second query. When the second
event fires, the associated script should cause the results to be returned.
Thus there should be no assumption that the return value of the script
associated with the first event is the html result. Much better to use
something like the current Httpd_ReturnData mechanism.
