Re: [xml-dev] A new approach for SAX-Java Binding

[Santhosh T <santhosh.tekuri@gmail.com>]

> The problem with Santhosh's project is that it still
> revolves around the idea of building an in-memory representation.

the intendation is to create domain object like Employee, Company
are you referring these classes as "in-memory representation"

if not:
    I don't create any dom like struture internally. the `SaxContext` which i am 
referring in wiki, is not same as dom like structure.

during parsing, the jlibs implementation will create only that many instances of SAXContext
which is equalent to maximum element depth in given document.

for each element, a SAX-Context is not created. as soon as an element ends, it's SAXContext
object is reused for the next sibling element.

- Santhosh

On Mon, Jan 25, 2010 at 3:24 PM, Boris Kolpackov <boris@codesynthesis.com> wrote:

Hi Michael,

Michael Kay <mike@saxonica.com> writes:

> > This idea might be new to Java..
>
> I can't see anything new about it. I can't see what it offers that all
> the other Java-XML data binding frameworks don't.

Most data binding tools convert XML data to an in-memory object model.
In other words, they are like statically-typed and named DOM. However,
this is not always desirable (streaming) or possible (large documents).
In such situations the "statically-typed and named SAX" approach is a
good  alternative. The problem with Santhosh's project is that it still
revolves around the idea of building an in-memory representation.

> and once you're there, it's a very small step to XSLT...

Oh please, not this discussion again ;-).

Boris

--
Boris Kolpackov, Code Synthesis        http://codesynthesis.com/~boris/blog
Open-source XML data binding for C++   http://codesynthesis.com/products/xsd
XML data binding for embedded systems  http://codesynthesis.com/products/xsde
Command line interface to C++ compiler http://codesynthesis.com/projects/cli