On Tue, Jun 27, 2000 at 06:08:29PM +0200, Kaiserovi wrote: > how do you generate the bugs, for God's sake? Are you running all the > possible ASCII sequences as stylesheets? I came across the first two working on a stylesheet, and the third one whilst trying to create a minimal test case for the second. I suspect I may be using XSLT in a rather unusual way (heavy use of mutual recursion). By the way, I've noticed that Sablotron allows you to use a result tree fragment exactly as if it were a node-set, i.e. you can write "$foo/*" where foo is bound to a result tree fragment. This appears to be in contravention of the spec, though it's incredibly useful. Other XSLT processors (XT, SAXON) provide an extension function "node-set" which can be used to coerce a result tree fragment to a node-set for this sort of thing. But of course they use different extension namespaces so you can't write a portable stylesheet using this. Is that a deliberate design decision, or a side-effect of the implementation? It would be lovely if there were a portable syntax for this powerful feature. (The bit of the spec I'm thinking of says "In particular, it is not permitted to use the /, //, and [] operators on result tree fragments." http://www.w3.org/TR/xslt#section-Result-Tree-Fragments The description of the "node-set" function in XT is at http://www.jclark.com/xml/xt.html#node-set ) .robin. -- Flee to me, remote elf!
