Re: Transitive closure for XPath

> If functions can be supplied as arguments to a function call or returned
as
> results of a function call, then it's reasonable to define a function that
> takes a function as its input and returns its transitive closure as its
> result; without that, it seems very difficult.

Indeed, that would be a nice way of going about it, although TC itself could
probably be achieved more cheaply..

For higher order functions, some lexical construct is needed that prevents
an XPath processor from evaluating the parameters passed to a function
before evaluating the function. If you want to pass a function as a
parameter, you will have to prevent the processor from evaluating that
function before passing it.

Christian Nentwich

Received on Tuesday, 24 April 2001 00:25:12 UTC