Re: XML-SAX *XML_EXCEPTION event and reported parsed length question

Hi Nathan

I believe that we don't tell the SAX parser where to go get the
information - it goes from start to end of the XML "document" and returns
to a call-out procedure with various events, like start of document, end of
element, stuff like that.

I understand that you don't tell XML-SAX which portions of an XML document
to parse. You specify the name of the procedure that the parser should call
when parsing events occur (i.e. when elements and attributes are found).

I was noting the difference with XMLTABLE, where you specify the name of
the XML element (which is normally an array of XML objects) to generate an
SQL cursor from.

You don't invoke the parser repetitively - at least, not the SAX parser in
use with XML-SAX operation code.

That's right. Again I was noting the difference between parsers like
XML-SAX or Expat vs. XMLTABLE. You would call XMLTABLE repetitively, once
for each cursor that you want to generate, from each XML array that you
specify.

As I indicated before, single-pass parsing with XML-SAX or Expat would
probably run a lot more efficiently than the repetitive passes that one
would invoke when using XMLTABLE.

A programmer has to know the structure ahead of time, in order to create
reasonable sets of procedures to process each event type. And you need to
keep track of where you are in the tree as the parser calls your procedure.

This IS a single pass - just lots of exits to the handling procedure you
specify.

Hope that makes sense and is even close to reality!!

Yes, I understand what you're saying. And hopefully you understand the
differences between XML-SAX and XMLTABLE.