[uf-discuss] Re: Optimus 0.5.1

[Toby A Inkster]

Michael MD wrote:

> nested includes? !
> my guess is that anyone doing that would be asking for trouble!

This is another thing that Cognition has supported since alpha1. (Though 
until the latest release it supported including a node's own parent, which 
really *is* asking for trouble!)

The only thing you've got to be careful of (from a parser's POV) is making 
sure that you don't get stuck in an infinite loop. The solution is to 
write your inclusion code to *not* support nested includes, and then 
simply call the function a few times. (The first call will handle 
includes, then the second call will handle includes within includes, etc.)

My policy is to allow two levels of includes for adr and geo, 4 levels for 
hCalendar valarm and hCalendar vfreebusy, and 6 for hCard, hAtom entries, 
hCalendar vevent and hCalendar todo.

I also use an optimisation such that each call of the function actually 
checks to see if any changes have been made. If no changes have been made, 
then the loop is ended prematurely -- this prevents the inclusion code 
(which is computationally expensive) from being called when there are no 
instances of class="include" left, or those instances are invalid (e.g. 
attempts to include an ancestor node).

-- 
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.17.14-mm-desktop-9mdvsmp, up 14 days, 19:59.]

                   Tagliatelle with Fennel and Asparagus
   http://tobyinkster.co.uk/blog/2008/04/06/tagliatelle-fennel-asparagus/