Re: Extension candidates

Michael Kay wrote:

> Then we can have a function ref($node) that returns an anonymous
> zero-argument function which, when called, yields $node.

> [...]

> So I'm inclined to build "nested sequences" from references as the
> primitive.

  I personally really like that idea.

  That brings the question: what would be the type of a function?  If I
am right, it would be a new specialized kind of item (as nodes are). 
If we were specifying XPath, we would add a new ItemType function(),
but for EXSLT we would just say an item() (that is not a node() nor any
defined simple type).

  As this is not a node, we couldn't use directly the same clever trick
used in FXSL: applying templates on the function object (required to be
a node,) and we would need ex:apply() to be an extension function.

  Regards,

-- 
Florent Georges
http://www.fgeorges.org/