On 04/01/2008, Colin Paul Adams <colin@xxxxxxxxxxxxxxxxxx> wrote:
> >>>>> "Andrew" == Andrew Welch <andrew.j.welch@xxxxxxxxx> writes:
>
>     Andrew> On 04/01/2008, Colin Paul Adams <colin@xxxxxxxxxxxxxxxxxx> wrote:
>     >> If you DIDN'T access the variable within the transformation at
>     >> all, when should the error be raised? Prior to the execution of
>     >> the initial template? That would be a bit harsh.
>
>     Andrew> Not sure if it aids the thread, but Saxon will throw an
>     Andrew> error for:
>
>     Andrew> <xsl:variable name="foo" as="element()"> <xsl:copy-of
>     Andrew> select="//@foo"/> </..
>
>     Andrew> ...using static analysis alone - regardless of whether the
>     Andrew> variable is used.
>
> You mean a static error at compile time? I don't suppose Florent
> intended ex:error-safe to be able to catch static errors. If he did,
> he will be severely disappointed.

I think Florent's well aware of most things - he contributes to Kernow
and his code is much better than mine :)

I was just replying to this:

    >> If you DIDN'T access the variable within the transformation at
   >> all, when should the error be raised? Prior to the execution of
   >> the initial template? That would be a bit harsh.

Mostly the error wouldn't get raised and the transform would complete,
but if it's possible to detect statically then the transform won't
compile (in Saxon anyway).


-- 
Andrew Welch
http://andrewjwelch.com
Kernow: http://kernowforsaxon.sf.net/