Tim Williams wrote:
> Sorry, I should have put my opinion in while I had the opportunity in
> your other thread on this. I think that moving some of these is good,
> but not all. For example, the extparser transformer doesn't seem to
> be used elsewhere so it makes sense to me to only have it defined in
> forrest.xmap. That way, if the requirement for it in forrest.xmap
> goes away we can confidently remove it from there. Otherwise, we
> won't easily know if theres dependencies elsewhere on it so it would
> sit forever. It follows the "define it as close to first use" idea.
Thanks for pointing that out.
I agree with your approach. Feel free to move back the stuff that is
unique to the forrest.xmap.
--
Ferdinand Soethe
|