Re: [xsl] My XPath mistakenly referenced an element that doesn't exist and I got no error message ... is this bad language design?

["Graydon graydon@xxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

On Thu, Oct 14, 2021 at 02:19:47PM -0000, Wendell Piez wapiez@xxxxxxxxxxxxxxx scripsit:
> Is this solution too demanding? (Not just defensive XSLT with extra tests in
> your XPath but an entire Schematron?) I always wonder, when we get to that
> point. What problem are we actually trying to solve?

We're trying to solve (at least) three:

- transform this to that
- ensure it's the expected this
- ensure it's the expected that

I think it's good practice to build these separately, rather than trying
to stuff the "expected this" part into the XSLT which is transforming
this to that.  (Which means failed "expected this" tests prevent the
transform from being attempted.)

I'd also suggest XQuery in preference to schematron for the "expected
this?" tests.  More power and available in "per collection" (and thus
"entire input content set") aware flavours.


-- 
Graydon Saunders  | graydonish@xxxxxxxxx
^fs oferiode, pisses swa mfg.
-- Deor  ("That passed, so may this.")