Re: [xsl] Confusing trace from Saxon HE9
Subject: Re: [xsl] Confusing trace from Saxon HE9|
From: Jeroen Hellingman <jeroen@xxxxxxxx>
Date: Wed, 09 Dec 2009 10:52:38 +0100
Thanks Michael for your explanation of tail recursion optimization.
Using the trace option helped me locate the real culprit, some template
in another style sheet fragment feeding
the splitter.xsl something it was not supposed to eat.
I will look into the possibility of declaring types for parameters
throughout my code. In this particular instance,
the $node parameter could be really any element from the source document
(but not a sequence), because what
I want to know is "in what output file fragment is this node rendered?",
so I can generate a link to it. This can
be quite tricky, considering the way I shuffle things around.
I guess I have to get the latest version of your book, as I am still
using the second edition...