RE: XSLT 2: 2.4 Executing a Transformation

David Carlisle:

> 2.4.1 says:
> 
> When stylesheet execution starts, a document node is
> implicitly created for the result tree 
> 
> 
> Which more or less matches the XSLT 1 behaviour.
> But is that now the right model? In particular it seems to 
> imply that the principal result tree is never empty.
> 
> If one arranges that the initial template just generates 
> secondary result trees it is in fact often useful not to 
> generate any principal output at all.
> 
> Couldn't the implicit creation of a root node be fired when 
> something is inserted into the principal result tree rather 
> than by stylesheet execution? 

This proposal is almost identical to one that the WG is currently
considering.

Thanks for the input.

Michael Kay

Received on Wednesday, 4 September 2002 15:46:00 UTC