This example demonstrates how to generate XML from non-XML data
sources.  This example is based directly on an example presented by
Tom Gavin and Joseph E. Hughes at the August 1999 Washington DC
SGML/XML User's Group meeting.  PowerPoint slides containing the
original DOM-based solution in Java are available at
http://www.eccnet.com/sgmlug/.

Since the specifics of reading other data formats vary greatly, this
example will use a simple comma-separated-value format similar to that
found as an "export" format for many applications which work with
tabular data.  A sample data file is contained in data.txt.

The loaddata.py script demonstrates three different approaches to XML
generation: DOM-based, SAX-based, and <file>.write()-based.  The first 
two approaches are specific to generating XML, while the third could
be used to generate any format.  It is interesting to note the
differences in code size to get roughly the same output using each of
the three approaches.

The script's main() function does little but parse the command line,
selecting the processing class appropriately.  Processing consists of
instantiating the processing class and calling its run() method.
Concrete subclasses of the abstract processing class determine the
actual machinery used to create the XML output.