Namespaces and Parsers

Peter Murray-Rust (peter@ursus.demon.co.uk)
Thu, 28 May 1998 09:34:18


I have been re-reading the namespace draft (I am a fan of namespaces!) as a
prelude to making JUMBO2 namespace-aware. I note that they are very closely
bound to the parsing process (WD-xml-names-1998-03-27):
'[the WG intends that] this namespace facility should become an integral
part of some future version of the XML spec.'

I suspect that the namespace draft will harden up before a future version
of XML (comments from WG?) and if so there are some technical requirements
for parsers. The draft gives a number of validity conditions (e.g. the
namespace declarations must precede the external and internal subsets). I
suspect that this positional and other validation information will be lost
in the output of many parsers (including SAX-compliant ones).

Should the current generation of parsers address the *syntactic* aspects of
namespaces? i.e. should it be possible to use a parser to:
- recover the namespace declarations and their components
- be notified of namespace validation errors

If the answer is 'yes', should this be part of the next version of SAX? My
immediate feeling is that it shouldn't be difficult (the NS draft is
pretty straightforward syntactically) and it's highly desirable. A
processNamespace() callback would save application writers for yet another
place where bugs and misunderstandings could creep in.

P.

Peter Murray-Rust, Director Virtual School of Molecular Sciences, domestic
net connection
VSMS http://www.nottingham.ac.uk/vsms, Virtual Hyperglossary
http://www.venus.co.uk/vhg