[
Lists Home |
Date Index |
Thread Index
]
Dennis Sosnoski wrote:
> I'm well aware of Infoset-based encodings, since my
> own XBIS project (http://www.xbis.org) is one of
> them. I was curious how ASN.1 deals with data at
> the Infoset level, rather than schema-specific versions.
I'm not an expert in these matters, however, it appears to me
that the "Fast Infoset" approach is actually very similar to the
approach of XBIS. Basically, there is an ASN.1 schema that defines
tags for various elements of the Infoset (elements, attributes,
unparsed entities, processing instructions, comments, etc.) and there
is a mechanism to insert new text strings into a "dictionary" as you
pass through the document and replace them with integer "handles" like
in XBIS. You can find an early (and now out-of-date) discussion of the
approach in the OSS Nokalva paper that was submitted to the Binary XML
Workshop.[1] As I understand it, the current approach and schema is
much simpler and more efficient than what was presented to the
Workshop also, there is a good bit of effort being put into it to make
sure that the final schema works well with SAX...
BTW: For anyone interested in the mapping of XSD to ASN.1, the
standard is now available on the web[2].
bob wyman
[1] See: Appendix A:
http://www.w3.org/2003/08/binary-interchange-workshop/32-OSS-Nokalva-P
osition-Paper-updated.pdf
[2] X.694, ASN.1 encoding rules - mapping W3C XML schema definitions
into ASN.1:
http://www.itu.int/ITU-T/studygroups/com17/languages/X.694-0401.pdf
|
