> 
> The following transformation achieves the same (level values are monotonous
> but not strictly consecutive, container/item info not used) and at the same
> time is twice shorter and considerably less complex.
> 
> It is just 33 lines (compared to 66 lines) and consists of just two
> templates one of which has one parameter and the other no parameters
> (compared to three templates one of which takes two parameters and another
> which takes four parameters):
> 

Dimitre,

in my code I was trying to get rid of exponential computation times, so popular
for XLST grouping problem, but often inadequate. My template processes 100 lines 
in 2 seconds, yours in 9 seconds. The same 100 lines repeated twice take 4 seconds 
with my template, and 106 seconds with yours. 

Of course, it depends on the problem to solve, but a couple of hundreds of entries 
in a hierarchical table of contents is not uncommon.

David 

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list