Rob Davis-5 <[EMAIL PROTECTED]> wrote on 03/13/2008 05:38:41 AM:

>
> To solved this I've binned using batik's own parser and am using the DOM
> LSParser instead.

   Well really it's just that you have bypassed Batik's wrapper around
the parser so you can get some of the intermediate information.  Note
that you still don't get the DOCTYPE reflected in the DOM or anything
you are just 'saving' it from the parsing operation.

> The comments are still omitted, however but that is a minor issue.

   The comments (like the DOCTYPE) are omitted because there is no
place in the DOM to store them.  Since our parser returns a DOM there
is no way to return them to you.

   In anycase DOCTYPEs are on the way out, to be replaced with schema
tied to namespaces.

> But I think I'm onto a winner because with the DOM LSParser I've got
> full control and not things hidden away or over customized as it
> seems with the batik parser.

   Sure, you have also replaced 3 lines of code with ~100, so I hope
you get something in return ;).  It's worth noting that there is
no 'batik parser' we simply wrap a standard XML parser to make it
easier for folks to build an SVG DOM (we also take care of providing
known SVG DTD's locally etc.)

   One issue with your code is that you are build an LS DOM
rather than a Batik/SVG DOM.  This is potentially a performance
issue because batik must clone your document to render it (as it
needs the CSS and SVG DOM interfaces).  If you aren't going to
render the document then this isn't an issue.

Reply via email to