Re: [xml-dev] XQuery Puzzle

[David Carlisle <davidc@nag.co.uk>]

> Actually, the failure wasn't because they were parentless, it was because
> they weren't siblings of each other

well yes and no, that's one way of viewing the failure of the
preceding-axis but as you observed the first place where the xpath was
selecting nothing is in the use of the $set//gotcha that is already
empty without needing to see that the elements in the sequence are not
siblings. The main point is that gotcha isn't a descendant of $set
(which can happen in xpath 1, admittedly) but in this case that's becaus
eit isn't the child of anything (which can't happen in xpath 1)

But my main point was just that for this particular example, since the
input can't occur in the xpath 1 data model, you can't really ask if
xpath 1 or 2 behaviour is the same. I didn't mean to imply that there
are no _similar_ constructs in xpath1.

David