Re: Lifetime of XSLTProcessorImpl

On Monday, March 17, 2003, at 05:10  AM, Carsten Ziegeler wrote:
> I just looked at code of the excalibur xmlutil XSLTProcessorImpl.
> It seems that it doesn't implement any lifetime interfaces,
> so in fact for each lookup a new instance is created.

Only inside the ECM :)

> Is anything preventing us from making it Poolable?
> Or did I oversee something?

Nope, go ahead and add Poolable for proper behavior under ECM and thus 
Cocoon. I'm also using that component under Fortress where that 
behavior is specified in metadata rather than in a marker interface.
-pete