On Tue, 4 Mar 2003, Dimitre Novatchev wrote:
don't forget that it was something just like this that provoked
my original posting about what match=node() really meant, and i'm
still wading through the responses, trying to summarize them.
in a nutshell, if you use that transform, xsltproc and saxon
behave differently:
saxon will copy over comments and processing instructions,
xsltproc will not
so, remind me again -- which one is behaving correctly?
This is not so difficult even to guess -- once you know that this is the
"identity transformation".
Identity is identity -- it must copy everything.
It seems to me that an XSLT processor that does not perform correctly the
identity transform (no offense, but this is difficult to believe)
demonstrates a complete lack of any testing...
i'm thoroughly new at this, so i couldn't take a definitive
stand on this one way or the other, but from my readings
which inspired me to ask about this in the first place,
it sure didn't *look* like there was one obviously
correct definition of an identity transform. (witness
the number of people who chimed in on this discussion.)
i'm willing to believe that two different people interpreted
the specs two different ways, that's all. no big deal.
rday
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list