half-baked parsers vs binary XML

[Gabe Beged-Dov]

I have been thinking about optimal XML parsing, partly as a result of
the binary XML discussion. Right now the world of XML parsers is divided
into well-formedness and validating. Another type being discussed is
binary.

I'd like to propose another, the half-baked parser. This parser is
mentioned in the notes for section 5.1 of the annotated XML spec (not in
a positive light :-).

The half-baked parser can only process XML documents that don't have a
prologue. This makes its memory footprint and execution path much
smaller and faster respectively. Unfortunately, it isn't a legal XML
parser anymore.

This can be addressed by having a modular parser architecture that would
be optimistic and try the half-baked parser first. If it encountered a
prologue, it could load either a  WF parsing module or a validating
parsing module.

I think that a highly tuned half-baked parser in combination with an
optional  stream-oriented compression scheme would address many of the
concerns that something like binary XML is intended to deal with in both
the transmission, storage and execution speed dimensions.

A great discussion of modular layered parsing can be found on Simon St.
Laurent's web site (www.simonstl.com).

Gabe Beged-Dov
www.jfinity.com


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)