Re: Namespaces and ixml

Tomos Hillman <yamahito@gmail.com> writes:
> Just a note to reiterate what Norm and Steven have already said: the

I think you meant “Norm and Michael”, not Steven, unless I’m
misunderstanding in some fundamental way.

> method above absolutely must NOT work anyway as an XML output. The
> relevant wording of the spec is "Grammars must be written so that any
> serialization of a parse tree produced from the grammar is well-formed
> XML."

I’m just going to note in passing, since I’ve already started an email
to the list, that while it might technically be possible to split a hair
here and assert that we could produce well-formed XML but not namespace
well-formed XML, since we’re nearly a quarter century on from the first
namespaces Recommendation, I don’t think that could be done except in
jest.

                                        Be seeing you,
                                          norm

--
Norm Tovey-Walsh
Saxonica