The lack of a pure ruby XSLT engine is on occasions annoying, because
then I
find myself descending to the nether regions of java, PHP and friends
I was thinking about the viability of an XSLT doc A that matchs
against arbitrary XSLT B and emits ruby code that effectivly performs
the
same transformation as the original XSLT document B.
Eventually the XSLT A would be able to ‘self compile’ to some ruby code
that
would replace it. Ie read some XSLT and emit functionally equivilent
ruby
code.
Does this sound like a sensible approach to getting ‘pure ruby’ XSLT
functionality?