Re: XML* vs. eXpat

Adam,

Yes - XML-SAX uses the same XML parser as eXpat.

I guess it's just the XML-INTO and XML-SAX syntax I don't like - plus it
just seems weird to me that they decided to do this - I mean, all other
opcodes & BIF's are 'generic' - READ (from a file), WRITE (to a file) etc.
And then they have this XML stuff, with %HANDLER and %XML parameters... At
least I would have called the BIF's something like INTO and PARSE...

PARSE( <document/variable-parameter> : <handler-parameter> :
feedback/comms-parameter> );

Either way, it seems to muuch like a full API call wrapped up in a BIF.

Finally, there are still limitations with RPG that I would have preferred to
see addressed before they spent time doing XML implementations that are
already done very well by third-party companies (and in the case of SEP, for
free).

Maybe I'm just grumpy this morning.

Rory

On Mon, Jul 7, 2008 at 10:41 AM, Adam Glauser <adamglauser@xxxxxxxxxxxx>
wrote:

Rory Hewitt wrote:

I just prefer code like this (pseudo-code):

defineHandler(type:handlerpointer);
defineHandler(type:handlerpointer);
defineHandler(type:handlerpointer);
defineParsingOptions(options);
parseXml(xml);
which is pretty much what RPG-XML does (and I think what SEP does).

Yeah, that's pretty much how eXpat works too. I thought that XML-SAX
worked in essentially the same way too. Isn't the SAX part referring to
the same Simple API for XML model that eXpat uses?

I haven't actually used the XML* BIFs, as I haven't had any new XML to
handle since we got to V5R4. It certainly looks like XML-INTO should be
pretty slick for smaller, simpler documents.