Ooops...actually I don't know. I haven't used the flat file version for so long, I don't know.

It hijacks the call to SITE_html::Display and wraps it in a set of nested subroutines. Which means, it could probably do the same to the wrapper for &load_template, as long as the number of links sent in was controlled in some way.

The problem I see with flat-file links is that the way it builds the category pages of links. This would have to be shoe-horned in into the "spanning pages" routine, where in Links SQL NG it doesn't. That process was broken up into two proceses -- span pages and build page. So, the "build page" gets a statement handle to the # of links for that page, and builds them all. It then returns to the span pages routine, which determines if another page needs to be built or not. The "build page" rountine never really knows what page it's on, or how many pages to build. In previous versions (SQL included), that was handled much, much differently.

For some reason, I thought this was originally posted in the Links SQL Discussion area.


PUGDOG®
PUGDOG® Enterprises, Inc.
FAQ: http://pugdog.com/FAQ