Robert S. Koberg wrote: > For example: > - why is xsl:include and xsl:import bad -> how should it be done > - why is XPath's document() bad -> how should it be done
Guesses: - Check for changes of included and imported XSLT files was implemented only recently (in 2.0.3, don't try this with 2.0.2) - Using gets simply gets NPE for the second page hit in 2.0.2, (fixed in 2.0.3, at least for me) - Check for changes of documents referenced by document() still not implemented in 2.0.3. Actually, using document() in 2.0.3 results in an NPE because the CVS code fixing point one above cannot cope with this. > - how do you reuse XML data that can be used for tabs, the nav, > snailtrails and links throughout the content (so you basically have a > lookup table as opposed to hard coding links) Hehe. Of course, I use xsl:import and document() quite extensively... With the possibility to refer to a variety of generators through internal pipelines you get a lot of power. Caching is the only promlem, I declared everything trivially as unchanging and cacheable, changes currently require a server shutdown and clearing the work directory. I hope Cocoon will have improved in this area once the issue becomes pressing for me. J.Pietschmann --------------------------------------------------------------------- Please check that your question has not already been answered in the FAQ before posting. <http://xml.apache.org/cocoon/faq/index.html> To unsubscribe, e-mail: <[EMAIL PROTECTED]> For additional commands, e-mail: <[EMAIL PROTECTED]>
