[Xml-bin] An attempt to sort things out [long]

[Al Snell]

On Wed, 18 Apr 2001, Stefan Zier wrote:

> On the other hand the API specs themselves already pretty much provide that
> framework. Other than some tools to make things easier for developers,
> theres not much that can be added IMHO. What I do like about this thought is
> that we could actually create multiple different binary encodings of XML,
> each specialized for certain criteria (e.g. fast SAX parsing or random
> access).

Yes, definitely. I see scope for at least three formats:

1) Basically packaged SAX events, for SAX-oriented applications that
rarely build DOM trees, or when streaming (as in transmitting the
document as it is constructed) is required.

> Maybe we could come up with some kind of envelope to wrap any binary data
> that identifies the specific parser to use. So for example the framework
> could pick up an IIOP message wrapped in this envelope and find the
> apropriate parser to create the SAX events the application is so desperate
> to get. Does this make any sense to anybody? Or do I probably just need more
> caffeine? *g*

I was thinking about:

<?xml version='1.0' ?>
<blah>
</blah>

...

<?xml version='1.0' format='SAXbin' format-version='0.5b' ?>
[00][04]blah[01][00][00][00][01][02]

...

<?xml version='1.0' format='DOMbin' format-version='0.7.6' ?>
...

ABS

-- 
                               Alaric B. Snell
 http://www.alaric-snell.com/  http://RFC.net/  http://www.warhead.org.uk/
   Any sufficiently advanced technology can be emulated in software